test(api): pin the resume-rewind contract

TestAudioResumeRewind drives the whole surface against the shared dev
DB with the fresh per-test regular user (the live admin is never
touched; the fixture book lives in a test-named library that cascades
away): default 0, profile PUT "5" rides the chapters payload,
non-allow-listed values 400 without side effects, explicit "0"
resets everywhere, and an omitted JSON key leaves the column
untouched.

Changelog entry in docs/contract_pins.md (additive chapters key +
the profile field's string-shaped PUT, keep_screen_awake precedent)
and the get_profile.md example gains the preference keys.

Debug note for the next agent: the first runs here 401'd with
'Ghost session rejected' — NOT a harness or auth bug. The tests never
execute schema.sql; the shared dev DB gets its schema from the app
container's boot. The rebuilt tests image ran the new GetUser
(selecting audio_resume_rewind_seconds) against a DB whose schema
came from the old app image — every GetUser errored, and the
ghost-session guard turned that into 401s. Applying the schema.sql
ALTER to the dev DB (what the next app rebuild does anyway) turned
the suite green. Until make rebuild-app runs, the running app
container serves the old binary.

.gitignore

@@ -87,3 +87,4 @@ session-*.md
 # see docs/contributing/development.md → Media fixtures)
 internal/services/testdata/
 test-results
+.openchamber/

AGENTS.md

@@ -145,6 +145,17 @@ every user sees every library unless an admin hides one) is enforced by
   new pages/controls go into `tests/ui/COVERAGE.md`). Item-scoped
   gates may answer 404 as well as 403 (row invisible vs Go-level
   check — contract_pins 2026-10-29).
+- **Collections RLS (2026-10-31)**: `collections` and
+  `collection_items` are governed too — predicates
+  `app_can_read_collection` / `app_can_write_collection`, PURE over
+  their arguments (never query `collections`, so no policy recursion
+  under FORCE RLS). Reads: owner-or-admin; writes: owner, admin escape
+  for SYSTEM collections only. Future sharing = one EXISTS branch
+  inside the predicate functions, no policy migration. Registration
+  seeding declares the new user's identity (`rls.WithUser` in
+  `CreateDefaultCollectionsForUser` — the JWT does not exist yet);
+  test-harness writes to non-system collections must run as the row's
+  OWNER (`scopeExecAs`/`ownerSeedCtx`, not the admin handle).
 
 History: 2026-10-13 (`431fec0`) the series/media listing surfaces
 leaked hidden libraries; **2026-10-29 enforcement moved to RLS** — the

Dockerfile

@@ -41,12 +41,14 @@ RUN npm run build:ts
 # Declared after `COPY . .` so a version bump only invalidates this layer.
 # The release workflow passes the git tag, e.g. APP_VERSION=v1.5.
 ARG APP_VERSION=dev
+# Short commit hash of the built source (stale-container drift guard).
+ARG GIT_COMMIT=none
 
 # Build Go binary (cached unless Go files or generated code changes)
 RUN --mount=type=cache,target=/root/go/pkg/mod \
   --mount=type=cache,target=/root/.cache/go-build \
   CGO_ENABLED=0 GOOS=linux go build -installsuffix cgo \
-  -ldflags "-X bookhoard/internal/version.Version=${APP_VERSION}" \
+  -ldflags "-X bookhoard/internal/version.Version=${APP_VERSION} -X bookhoard/internal/version.Commit=${GIT_COMMIT}" \
   -o main ./cmd/server
 
 # Materialize the Go module cache into an image layer so the test-runner

Makefile

@@ -11,6 +11,10 @@ endif
 # Override with: CONTAINER_RUNTIME=podman make rebuild-app
 CONTAINER_RUNTIME ?= $(shell command -v docker 2>/dev/null || command -v podman 2>/dev/null)
 
+# Build stamp (stale-container drift guard): the rebuilt image carries
+# the source commit; the app logs it at boot and /api/health reports it.
+GIT_COMMIT ?= $(shell git rev-parse --short HEAD 2>/dev/null || echo none)
+
 # Dev compose stack: base prod file merged with the dev override (local build + tests).
 # Prod deploy does NOT use this — it runs plain `docker compose` against the base file only.
 COMPOSE := $(CONTAINER_RUNTIME) compose -f docker-compose.yml -f docker-compose.dev.yml

bruno/collections/add_auto_assign_rule.yml

@@ -1,41 +0,0 @@
-info:
-  name: Add Auto-Assign Rule
-  type: http
-  seq: 1
-
-http:
-  method: POST
-  url: '{{base_url}}/api/collections/{{collection_id}}/rules'
-  auth: inherit
-  body:
-    type: json
-    data: |-
-      {
-        "field": "genre",
-        "operator": "equals",
-        "value": "Science Fiction",
-        "priority": 1,
-        "enabled": true
-      }
-settings:
-  encodeUrl: true
-  timeout: 0
-  followRedirects: true
-  maxRedirects: 5
-
-runtime:
-  scripts:
-    - type: tests
-      code: |-
-        test("status is 201", function() {
-          expect(res.status).to.eql(201);
-        });
-
-docs:
-  content: |-
-    # Add Auto-Assign Rule
-
-    `POST /api/collections/:id/rules` — auth: jwt.
-
-    Canonical reference: docs/developer/api/collections/add_auto_assign_rule.md (docs site → Developer → API). This request is GENERATED from that doc; edit the doc, then run the drift test with -update.
-  type: text/markdown

bruno/collections/create_collection.yml

@@ -16,7 +16,6 @@ http:
         "color": "#00FF00",
         "icon": "📖",
         "collection_type": "static",
-        "auto_assign_rules": [],
         "view_settings": { "kobo": { "view_mode": "grid" } }
       }
 settings:

bruno/collections/remove_auto_assign_rule.yml

@@ -1,31 +0,0 @@
-info:
-  name: Remove Auto-Assign Rule
-  type: http
-  seq: 13
-
-http:
-  method: DELETE
-  url: '{{base_url}}/api/collections/{{collection_id}}/rules/{{rule_id}}'
-  auth: inherit
-settings:
-  encodeUrl: true
-  timeout: 0
-  followRedirects: true
-  maxRedirects: 5
-
-runtime:
-  scripts:
-    - type: tests
-      code: |-
-        test("status is 204", function() {
-          expect(res.status).to.eql(204);
-        });
-
-docs:
-  content: |-
-    # Remove Auto-Assign Rule
-
-    `DELETE /api/collections/:id/rules/:ruleId` — auth: jwt.
-
-    Canonical reference: docs/developer/api/collections/remove_auto_assign_rule.md (docs site → Developer → API). This request is GENERATED from that doc; edit the doc, then run the drift test with -update.
-  type: text/markdown

bruno/collections/set_chain_order.yml

@@ -0,0 +1,29 @@
+info:
+  name: Set Collection Chain Order
+  type: http
+  seq: 13
+
+http:
+  method: PUT
+  url: '{{base_url}}/api/collections/{{collection_id}}/view-settings'
+  auth: inherit
+  body:
+    type: json
+    data: |-
+      {
+        "chain_order": "newest"
+      }
+settings:
+  encodeUrl: true
+  timeout: 0
+  followRedirects: true
+  maxRedirects: 5
+
+docs:
+  content: |-
+    # Set Collection Chain Order
+
+    `PUT /api/collections/:id/view-settings` — auth: jwt.
+
+    Canonical reference: docs/developer/api/collections/set_chain_order.md (docs site → Developer → API). This request is GENERATED from that doc; edit the doc, then run the drift test with -update.
+  type: text/markdown

bruno/collections/test_rule.yml

@@ -1,48 +0,0 @@
-info:
-  name: Test Rule
-  type: http
-  seq: 15
-
-http:
-  method: POST
-  url: '{{base_url}}/api/collections/test-rules'
-  auth: inherit
-  body:
-    type: json
-    data: |-
-      {
-        "rules": [
-          {
-            "field": "genre",
-            "operator": "equals",
-            "value": "Science Fiction"
-          },
-          {
-            "field": "author",
-            "operator": "contains",
-            "value": "Asimov"
-          }
-        ]
-      }
-settings:
-  encodeUrl: true
-  timeout: 0
-  followRedirects: true
-  maxRedirects: 5
-
-runtime:
-  scripts:
-    - type: tests
-      code: |-
-        test("status is 200", function() {
-          expect(res.status).to.eql(200);
-        });
-
-docs:
-  content: |-
-    # Test Rule
-
-    `POST /api/collections/test-rules` — auth: jwt.
-
-    Canonical reference: docs/developer/api/collections/test_rule.md (docs site → Developer → API). This request is GENERATED from that doc; edit the doc, then run the drift test with -update.
-  type: text/markdown

bruno/collections/update_collection.yml

@@ -15,7 +15,6 @@ http:
         "description": "Updated description",
         "color": "#33CCFF",
         "icon": "📚",
-        "auto_assign_rules": [],
         "view_settings": { "hide_finished": true },
         "hide_finished": ["true"],
         "show_up_next": ["true"]

bruno/collections/update_rule.yml

@@ -1,29 +0,0 @@
-info:
-  name: Update Collection Rule
-  type: http
-  seq: 17
-
-http:
-  method: PUT
-  url: '{{base_url}}/api/collections/{{collection_id}}/rules/{{rule_id}}'
-  auth: inherit
-  body:
-    type: json
-    data: |-
-      {
-        "enabled": false
-      }
-settings:
-  encodeUrl: true
-  timeout: 0
-  followRedirects: true
-  maxRedirects: 5
-
-docs:
-  content: |-
-    # Update Collection Rule
-
-    `PUT /api/collections/:id/rules/:ruleId` — auth: jwt.
-
-    Canonical reference: docs/developer/api/collections/update_rule.md (docs site → Developer → API). This request is GENERATED from that doc; edit the doc, then run the drift test with -update.
-  type: text/markdown

bruno/environments/Bookhoard.yml

@@ -60,3 +60,13 @@ variables:
     name: book_sha256
   - secret: true
     name: book_uuid
+  - secret: true
+    name: podcasts_library_id
+  - name: podcast_feed_url
+    value: https://example.com/feed.xml
+  - secret: true
+    name: podcast_feed_id
+  - secret: true
+    name: podcast_episode_id
+  - name: podcast_search_query
+    value: episode

bruno/podcasts/add_feed.yml

@@ -0,0 +1,30 @@
+info:
+  name: Add Podcast Feed
+  type: http
+  seq: 1
+
+http:
+  method: POST
+  url: '{{base_url}}/api/podcasts/feeds'
+  auth: inherit
+  body:
+    type: json
+    data: |-
+      {
+        "library_id": "{{podcasts_library_id}}",
+        "url": "{{podcast_feed_url}}",
+        "download_all": false
+      }
+settings:
+  encodeUrl: true
+  timeout: 30000
+  followRedirects: true
+  maxRedirects: 5
+
+runtime:
+  scripts:
+    - type: tests
+      code: |-
+        test("status is 201", function() {
+          expect(res.status).to.eql(201);
+        });

bruno/podcasts/delete_feed.yml

@@ -0,0 +1,17 @@
+info:
+  name: Delete Feed (admin; ?files=false keeps files)
+  type: http
+  seq: 12
+
+http:
+  method: DELETE
+  url: '{{base_url}}/api/podcasts/feeds/{{podcast_feed_id}}'
+  auth: inherit
+
+runtime:
+  scripts:
+    - type: tests
+      code: |-
+        test("status is 200", function() {
+          expect(res.status).to.eql(200);
+        });

bruno/podcasts/download_all.yml

@@ -0,0 +1,17 @@
+info:
+  name: Download All (every non-downloaded episode)
+  type: http
+  seq: 8
+
+http:
+  method: POST
+  url: '{{base_url}}/api/podcasts/feeds/{{podcast_feed_id}}/episodes/all/download'
+  auth: inherit
+
+runtime:
+  scripts:
+    - type: tests
+      code: |-
+        test("status is 202", function() {
+          expect(res.status).to.eql(202);
+        });

bruno/podcasts/download_episode.yml

@@ -0,0 +1,17 @@
+info:
+  name: Download / Retry Episode (idempotent state machine)
+  type: http
+  seq: 7
+
+http:
+  method: POST
+  url: '{{base_url}}/api/podcasts/feeds/{{podcast_feed_id}}/episodes/{{podcast_episode_id}}/download'
+  auth: inherit
+
+runtime:
+  scripts:
+    - type: tests
+      code: |-
+        test("status is 202", function() {
+          expect(res.status).to.eql(202);
+        });

bruno/podcasts/get_feed.yml

@@ -0,0 +1,17 @@
+info:
+  name: Feed Detail (episodes + download states)
+  type: http
+  seq: 3
+
+http:
+  method: GET
+  url: '{{base_url}}/api/podcasts/feeds/{{podcast_feed_id}}'
+  auth: inherit
+
+runtime:
+  scripts:
+    - type: tests
+      code: |-
+        test("status is 200 (404 would be scoping/RLS)", function() {
+          expect(res.status).to.eql(200);
+        });

bruno/podcasts/list_feeds.yml

@@ -0,0 +1,17 @@
+info:
+  name: List My Podcast Feeds
+  type: http
+  seq: 2
+
+http:
+  method: GET
+  url: '{{base_url}}/api/podcasts/feeds'
+  auth: inherit
+
+runtime:
+  scripts:
+    - type: tests
+      code: |-
+        test("status is 200", function() {
+          expect(res.status).to.eql(200);
+        });

bruno/podcasts/refresh_feed.yml

@@ -0,0 +1,17 @@
+info:
+  name: Manual Refresh (poll one feed)
+  type: http
+  seq: 6
+
+http:
+  method: POST
+  url: '{{base_url}}/api/podcasts/feeds/{{podcast_feed_id}}/refresh'
+  auth: inherit
+
+runtime:
+  scripts:
+    - type: tests
+      code: |-
+        test("status is 200", function() {
+          expect(res.status).to.eql(200);
+        });

bruno/podcasts/sequence.yml

@@ -0,0 +1,17 @@
+info:
+  name: Effective Sequence (override applied, downloaded-only, gaps)
+  type: http
+  seq: 9
+
+http:
+  method: GET
+  url: '{{base_url}}/api/podcasts/feeds/{{podcast_feed_id}}/sequence'
+  auth: inherit
+
+runtime:
+  scripts:
+    - type: tests
+      code: |-
+        test("status is 200", function() {
+          expect(res.status).to.eql(200);
+        });

bruno/podcasts/set_order.yml

@@ -0,0 +1,23 @@
+info:
+  name: Set Order Override (native | oldest | newest)
+  type: http
+  seq: 10
+
+http:
+  method: PUT
+  url: '{{base_url}}/api/podcasts/feeds/{{podcast_feed_id}}/order'
+  auth: inherit
+  body:
+    type: json
+    data: |-
+      {
+        "order": "oldest"
+      }
+
+runtime:
+  scripts:
+    - type: tests
+      code: |-
+        test("status is 200", function() {
+          expect(res.status).to.eql(200);
+        });

bruno/podcasts/subscription_search.yml

@@ -0,0 +1,17 @@
+info:
+  name: Subscription-scoped Search (covers un-downloaded + historical)
+  type: http
+  seq: 11
+
+http:
+  method: GET
+  url: '{{base_url}}/api/podcasts/search?q={{podcast_search_query}}'
+  auth: inherit
+
+runtime:
+  scripts:
+    - type: tests
+      code: |-
+        test("status is 200", function() {
+          expect(res.status).to.eql(200);
+        });

bruno/podcasts/unsubscribe.yml

@@ -0,0 +1,17 @@
+info:
+  name: Unsubscribe (files retained)
+  type: http
+  seq: 5
+
+http:
+  method: DELETE
+  url: '{{base_url}}/api/podcasts/feeds/{{podcast_feed_id}}/subscription'
+  auth: inherit
+
+runtime:
+  scripts:
+    - type: tests
+      code: |-
+        test("status is 200", function() {
+          expect(res.status).to.eql(200);
+        });

bruno/podcasts/update_subscription.yml

@@ -0,0 +1,25 @@
+info:
+  name: Update Subscription (display name + auto-download)
+  type: http
+  seq: 4
+
+http:
+  method: PUT
+  url: '{{base_url}}/api/podcasts/feeds/{{podcast_feed_id}}/subscription'
+  auth: inherit
+  body:
+    type: json
+    data: |-
+      {
+        "display_name": "My Serialized Feed",
+        "auto_download": "newest",
+        "auto_download_count": 3
+      }
+
+runtime:
+  scripts:
+    - type: tests
+      code: |-
+        test("status is 200", function() {
+          expect(res.status).to.eql(200);
+        });

cmd/server/main.go

@@ -10,6 +10,7 @@ import (
 	"bookhoard/internal/router"
 	"bookhoard/internal/services"
 	"bookhoard/internal/sync"
+	"bookhoard/internal/version"
 	"bookhoard/templates"
 	"context"
 	"log"
@@ -33,6 +34,7 @@ func (cv *CustomValidator) Validate(i interface{}) error {
 }
 
 func main() {
+	log.Printf("Bookhoard %s (commit %s) starting", version.Version, version.Commit)
 	cfg := config.LoadConfig()
 
 	dbPool, err := pgxpool.New(context.Background(), cfg.DatabaseURL())
@@ -168,15 +170,27 @@ func main() {
 	opdsHandler := handlers.NewOPDSHandler(queries, libraryService, conversionService)
 	opdsHandler.SetSettings(registry)
 
-	collectionHandler := handlers.NewCollectionHandler(queries, libraryService, connManager)
+	// One composed-engine service shared by search AND the collection
-	dashboardService := services.NewDashboardService(queries)
+	// subsystem (item 3.2): tree-driven smart collections evaluate
-	dashboardHandler := handlers.NewDashboardHandler(queries)
+	// through the unified engine.
+	collectionSearchService := services.NewSearchService(queries, scopedDB)
+	collectionHandler := handlers.NewCollectionHandler(queries, libraryService, connManager, collectionSearchService)
+	dashboardService := services.NewDashboardService(queries, collectionSearchService)
+	dashboardHandler := handlers.NewDashboardHandler(queries, collectionSearchService)
 	seriesHandler := handlers.NewSeriesHandler(queries)
 	filtersHandler := handlers.NewFiltersHandler(queries)
 	mediaHandler := handlers.NewMediaHandler(queries, scopedDB, libraryService, progressService, annotationService)
 	matchingHandler := handlers.NewMatchingHandler(queries, connManager)
 	jobsHandler := handlers.NewJobsHandler(queries, worker)
 
+	// Podcast subsystem: fetcher service + concurrent download worker
+	// (§0 item 4: pool, fair share, admin tunables) + poller. One
+	// instance trio per server (per-feed locks dedupe).
+	podcastService := services.NewPodcastService(queries)
+	podcastWorker := services.NewPodcastDownloadWorker(queries, podcastService)
+	podcastWorker.SetSettings(registry)
+	podcastHandler := handlers.NewPodcastHandler(queries, podcastService, podcastWorker)
+
 	e := echo.New()
 
 	// Set up validator
@@ -237,6 +251,9 @@ func main() {
 		CollectionHandler:       collectionHandler,
 		FiltersHandler:          filtersHandler,
 		DashboardHandler:        dashboardHandler,
+		PodcastHandler:          podcastHandler,
+		PodcastService:          podcastService,
+		PodcastWorker:           podcastWorker,
 		DashboardService:        dashboardService,
 		SeriesHandler:           seriesHandler,
 		OPDSHandler:             opdsHandler,

cmd/server/tests/audiobook_multifile_test.go

@@ -659,12 +659,14 @@ func TestAudiobookFolderSidecarEditDetected(t *testing.T) {
 	require.NoError(t, os.WriteFile(sidecar, []byte(`{"title":"Sidecar Real Titlz","series":["Sidecar Saga (Reading Order) #1"]}`), 0o644))
 	scanTestFolder(t, setup, admin, folder)
 	var collCount int
-	require.NoError(t, setup.DBPool.QueryRow(context.Background(),
+	// Owner-scoped read: collections RLS (2026-10-31) hides rows from
+	// the pool session's SeedIdentity.
+	require.NoError(t, database.NewScopedDB(setup.DBPool).QueryRow(ownerSeedCtx(admin),
 		`SELECT COUNT(*) FROM collections WHERE name = 'Sidecar Saga (Reading Order)' AND user_id = $1`, admin,
 	).Scan(&collCount))
 	require.Equal(t, 1, collCount, "reading-order sidecar edit creates the collection for an existing book")
 	scanTestFolder(t, setup, admin, folder)
-	require.NoError(t, setup.DBPool.QueryRow(context.Background(),
+	require.NoError(t, database.NewScopedDB(setup.DBPool).QueryRow(ownerSeedCtx(admin),
 		`SELECT COUNT(*) FROM collections WHERE name = 'Sidecar Saga (Reading Order)' AND user_id = $1`, admin,
 	).Scan(&collCount))
 	require.Equal(t, 1, collCount, "rescan is idempotent — no duplicate collection")

cmd/server/tests/collections_rls_test.go

@@ -0,0 +1,127 @@
+package main
+
+import (
+	"bytes"
+	"testing"
+	"encoding/json"
+	"fmt"
+	"net/http"
+	"time"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+// Collections RLS (2026-10-31): ownership enforcement moved to the DB
+// layer (p_collections_* + parent-checked collection_items policies in
+// schema.sql, predicates app_can_read/write_collection). The API surface
+// answers invisibility with 404 — including the write endpoints that
+// had NO app-level gate (Update/AddBooks, the flagged gap). Admin keeps
+// the escape; system collections stay per-user; registration still
+// seeds them (the seeding identity fix in
+// CreateDefaultCollectionsForUser is regression-pinned here).
+func TestCollectionsRLS(t *testing.T) {
+	setup := setupTestServer(t)
+	ownerToken := setup.RegularToken // user A: the harness regular user
+	adminToken := setup.Token
+	marker := fmt.Sprintf("rls%d", time.Now().UnixNano()%100000)
+
+	// User B: a second regular user via registration. The register
+	// endpoint is rate-limited — exactly one call. The response's
+	// access_token is B's identity for every probe below.
+	regEmail := fmt.Sprintf("%s@tests.bookhoard.internal", marker)
+	regBody := fmt.Sprintf(
+		`{"email":%q,"username":%q,"password":"Test@Pass123!"}`,
+		regEmail, "rlsb"+marker)
+	regResp, err := http.Post(setup.Server.URL+"/api/auth/register",
+		"application/json", bytes.NewReader([]byte(regBody)))
+	require.NoError(t, err)
+	require.Equal(t, http.StatusCreated, regResp.StatusCode,
+		"registration must succeed — including the system-collection seeding that now runs under the new user's RLS identity")
+	var reg struct {
+		Token string `json:"access_token"`
+	}
+	require.NoError(t, json.NewDecoder(regResp.Body).Decode(&reg))
+	require.NotEmpty(t, reg.Token)
+	otherToken := reg.Token
+	_ = regResp.Body.Close()
+
+	// B's registration seeded B's OWN system collections — visible in
+	// B's list (this is the registration-seeding identity fix, live).
+	code, body := podcastAuthed(t, setup, otherToken, "GET", "/api/collections", "")
+	require.Equal(t, http.StatusOK, code)
+	assert.Contains(t, string(body), "Continue Reading",
+		"registration seeded the new user's system collections under RLS")
+
+	// A creates the target collection.
+	code, body = podcastAuthed(t, setup, ownerToken, "POST", "/api/collections",
+		`{"name":"test RLS Owner ` + marker + `"}`)
+	require.Equal(t, http.StatusCreated, code, "%s", string(body))
+	var coll struct {
+		ID string `json:"id"`
+	}
+	require.NoError(t, json.Unmarshal(body, &coll))
+	t.Cleanup(func() {
+		podcastAuthed(t, setup, ownerToken, "DELETE", "/api/collections/"+coll.ID, "")
+	})
+
+	// ── B (non-owner) probes A's collection: invisible = 404 ──
+	code, _ = podcastAuthed(t, setup, otherToken, "GET", "/api/collections/"+coll.ID, "")
+	require.Equal(t, http.StatusNotFound, code, "detail-by-id is RLS-invisible cross-user")
+
+	code, body = podcastAuthed(t, setup, otherToken, "GET", "/api/collections", "")
+	require.Equal(t, http.StatusOK, code)
+	assert.NotContains(t, string(body), "test RLS Owner "+marker,
+		"the list query stays user-scoped")
+
+	code, body = podcastAuthed(t, setup, otherToken, "PUT", "/api/collections/"+coll.ID,
+		`{"name":"renamed by B"}`)
+	require.Equal(t, http.StatusNotFound, code,
+		"update of an RLS-invisible row answers 404 (was: ungated success): %s", string(body))
+
+	code, body = podcastAuthed(t, setup, otherToken, "POST", "/api/collections/"+coll.ID+"/books",
+		`{"book_ids":[]}`)
+	require.Equal(t, http.StatusNotFound, code,
+		"AddBooks on an RLS-invisible collection answers 404 (the flagged ungated-write gap, closed at the DB)")
+
+	code, body = podcastAuthed(t, setup, otherToken, "PUT",
+		"/api/collections/"+coll.ID+"/view-settings", `{"chain_order":"newest"}`)
+	require.Equal(t, http.StatusNotFound, code,
+		"chain-order write on an invisible collection answers 404")
+
+	code, _ = podcastAuthed(t, setup, otherToken, "DELETE", "/api/collections/"+coll.ID, "")
+	require.Equal(t, http.StatusNoContent, code,
+		"DELETE touches no rows under RLS (silent no-op — no destructive action)")
+
+	// The no-op delete lied about nothing: A still owns the collection.
+	code, body = podcastAuthed(t, setup, ownerToken, "GET", "/api/collections/"+coll.ID, "")
+	require.Equal(t, http.StatusOK, code, "owner retains the collection after B's failed delete")
+
+	// ── admin escape: sees everything, as with media ──
+	code, _ = podcastAuthed(t, setup, adminToken, "GET", "/api/collections/"+coll.ID, "")
+	require.Equal(t, http.StatusOK, code, "admin persona sees collections (policy escape)")
+
+	// ── B's ownership works: B renames B's OWN system collection ──
+	// (positive control — B is not locked out of what B owns)
+	code, body = podcastAuthed(t, setup, otherToken, "GET", "/api/collections", "")
+	require.Equal(t, http.StatusOK, code)
+	var list struct {
+		Collections []struct {
+			ID   string `json:"id"`
+			Name string `json:"name"`
+		} `json:"collections"`
+	}
+	require.NoError(t, json.Unmarshal(body, &list))
+	var bSystem string
+	for _, c := range list.Collections {
+		if c.Name == "Continue Reading" {
+			bSystem = c.ID
+			break
+		}
+	}
+	require.NotEmpty(t, bSystem, "B owns the seeded system collections")
+	code, body = podcastAuthed(t, setup, otherToken, "PUT", "/api/collections/"+bSystem,
+		`{"name":"Continue Reading"}`)
+	require.Contains(t, []int{http.StatusOK, http.StatusCreated}, code,
+		"owner mutation of an OWN collection still works: %s", string(body))
+}

cmd/server/tests/collections_smart_test.go

@@ -56,36 +56,6 @@ func smartGetCollection(t *testing.T, setup *TestServerSetup, token, collectionI
 	return result
 }
 
-func smartPostRule(t *testing.T, setup *TestServerSetup, token, collectionID, field, operator, value string, enabled *bool) *http.Response {
-	t.Helper()
-	payload := map[string]interface{}{
-		"field":    field,
-		"operator": operator,
-		"value":    value,
-	}
-	if enabled != nil {
-		payload["enabled"] = *enabled
-	}
-	body, _ := json.Marshal(payload)
-	req, _ := http.NewRequest("POST", setup.Server.URL+"/api/collections/"+collectionID+"/rules", bytes.NewBuffer(body))
-	req.Header.Set("Content-Type", "application/json")
-	req.Header.Set("Authorization", "Bearer "+token)
-	resp, err := http.DefaultClient.Do(req)
-	require.NoError(t, err)
-	return resp
-}
-
-func smartDecodeRuleID(t *testing.T, resp *http.Response) string {
-	t.Helper()
-	defer func(Body io.ReadCloser) { _ = Body.Close() }(resp.Body)
-	var result map[string]interface{}
-	require.NoError(t, json.NewDecoder(resp.Body).Decode(&result))
-	rule := result["rule"].(map[string]interface{})
-	return rule["id"].(string)
-}
-
-// Data helpers -----------------------------------------------------------
-
 func smartCreateLibraryAndBooks(t *testing.T, setup *TestServerSetup, userID uuid.UUID, titlesByGenre map[string]string) (libraryID string, bookIDs []string) {
 	t.Helper()
 	ctx := context.Background()
@@ -145,81 +115,83 @@ func TestSmartCollections(t *testing.T) {
 	setup := setupTestServer(t)
 	adminID := getTestUserID(t, setup.DB)
 
-	t.Run("FirstRuleConvertsCollectionToSmart_LiveMembership", func(t *testing.T) {
+	t.Run("TreeCollectionLiveMembership", func(t *testing.T) {
+		// The unified engine (item 3.2): a collection created WITH a
+		// books-scope tree evaluates LIVE — stored links are ignored,
+		// only tree matches render.
 		_, bookIDs := smartCreateLibraryAndBooks(t, setup, adminID, map[string]string{
 			"Smart SF Book One": "Science Fiction",
 			"Smart SF Book Two": "Science Fiction",
 			"Smart FANT Book":   "Fantasy",
 		})
+		_ = bookIDs
 
-		// Rule evaluation ignores stored links: a static collection's own
+		tree := map[string]interface{}{
-		// added item must NOT show up once the collection is smart.
+			"scope": "books", "mode": "all",
-		collID := smartCreateCollection(t, setup, setup.Token, "Test Smart Live "+uuid.NewString()[:8], "static")
+			"children": []map[string]interface{}{
-
+				{"field": "genre", "op": "is", "values": []string{"Science Fiction"}},
-		// Add one hand-picked book (a Fantasy one that the smart rule
+			},
-		// will not match — so it must vanish from the render).
+		}
-		addBody, _ := json.Marshal(map[string]interface{}{"book_ids": []string{bookIDs[2]}})
+		body, _ := json.Marshal(map[string]interface{}{
-		addReq, _ := http.NewRequest("POST", setup.Server.URL+"/api/collections/"+collID+"/books", bytes.NewBuffer(addBody))
+			"name":        "Test Smart Tree " + uuid.NewString()[:8],
-		addReq.Header.Set("Content-Type", "application/json")
+			"filter_tree": tree,
-		addReq.Header.Set("Authorization", "Bearer "+setup.Token)
+		})
-		addResp, err := http.DefaultClient.Do(addReq)
+		req, _ := http.NewRequest("POST", setup.Server.URL+"/api/collections", bytes.NewBuffer(body))
+		req.Header.Set("Content-Type", "application/json")
+		req.Header.Set("Authorization", "Bearer "+setup.Token)
+		resp, err := http.DefaultClient.Do(req)
 		require.NoError(t, err)
-		require.Equal(t, http.StatusNoContent, addResp.StatusCode)
-		_ = addResp.Body.Close()
-
-		staticDetail := smartGetCollection(t, setup, setup.Token, collID)
-		assert.Equal(t, false, staticDetail["is_smart"])
-		assert.Equal(t, float64(1), staticDetail["book_count"], "static collection shows its stored link")
-
-		resp := smartPostRule(t, setup, setup.Token, collID, "genre", "equals", "Science Fiction", nil)
 		require.Equal(t, http.StatusCreated, resp.StatusCode)
+		var created struct {
+			ID string `json:"id"`
+		}
+		require.NoError(t, json.NewDecoder(resp.Body).Decode(&created))
 		_ = resp.Body.Close()
+		t.Cleanup(func() {
+			del, _ := http.NewRequest("DELETE", setup.Server.URL+"/api/collections/"+created.ID, nil)
+			del.Header.Set("Authorization", "Bearer "+setup.Token)
+			_, _ = http.DefaultClient.Do(del)
+		})
 
-		smartDetail := smartGetCollection(t, setup, setup.Token, collID)
+		detail := smartGetCollection(t, setup, setup.Token, created.ID)
-		assert.Equal(t, true, smartDetail["is_smart"], "rules presence derives the type")
+		assert.Equal(t, true, detail["is_smart"], "tree presence derives the type")
-		assert.Equal(t, float64(2), smartDetail["book_count"], "live rule matches only")
+		memberTitles := smartLibraryTitleSet(t, detail)
-
-		memberTitles := smartLibraryTitleSet(t, smartDetail)
 		assert.True(t, memberTitles["Smart SF Book One"])
 		assert.True(t, memberTitles["Smart SF Book Two"])
-		assert.False(t, memberTitles["Smart FANT Book"], "hand-picked link is ignored under smart evaluation (v1 no hybrid)")
+		assert.False(t, memberTitles["Smart FANT Book"], "the tree's AND governs membership live")
-	})
-
-	t.Run("DisabledRuleExcludedFromLiveSet", func(t *testing.T) {
-		smartCreateLibraryAndBooks(t, setup, adminID, map[string]string{
-			"Smart Diss Book": "Science Fiction",
-		})
-		collID := smartCreateCollection(t, setup, setup.Token, "Test Smart Disabled "+uuid.NewString()[:8], "smart")
-
-		resp := smartPostRule(t, setup, setup.Token, collID, "genre", "equals", "Science Fiction", nil)
-		require.Equal(t, http.StatusCreated, resp.StatusCode)
-		ruleID := smartDecodeRuleID(t, resp)
-
-		assert.Equal(t, float64(1), smartGetCollection(t, setup, setup.Token, collID)["book_count"])
-
-		// Disable: membership drops to zero at once.
-		falseFlag := false
-		dBody, _ := json.Marshal(map[string]interface{}{"enabled": falseFlag})
-		dReq, _ := http.NewRequest("PUT", setup.Server.URL+"/api/collections/"+collID+"/rules/"+ruleID, bytes.NewBuffer(dBody))
-		dReq.Header.Set("Content-Type", "application/json")
-		dReq.Header.Set("Authorization", "Bearer "+setup.Token)
-		dResp, err := http.DefaultClient.Do(dReq)
-		require.NoError(t, err)
-		require.Equal(t, http.StatusOK, dResp.StatusCode)
-		_ = dResp.Body.Close()
-
-		assert.Equal(t, float64(0), smartGetCollection(t, setup, setup.Token, collID)["book_count"], "disabled rules never match")
 	})
 
 	t.Run("SmartCollectionsRejectMembershipMutations", func(t *testing.T) {
 		smartCreateLibraryAndBooks(t, setup, adminID, map[string]string{
 			"Smart Guard Book": "Science Fiction",
 		})
-		collID := smartCreateCollection(t, setup, setup.Token, "Test Smart Guard "+uuid.NewString()[:8], "smart")
+		tree := map[string]interface{}{
-
+			"scope": "books", "mode": "all",
-		resp := smartPostRule(t, setup, setup.Token, collID, "genre", "equals", "Science Fiction", nil)
+			"children": []map[string]interface{}{
+				{"field": "genre", "op": "is", "values": []string{"Science Fiction"}},
+			},
+		}
+		body, _ := json.Marshal(map[string]interface{}{
+			"name":        "Test Smart Guard " + uuid.NewString()[:8],
+			"filter_tree": tree,
+		})
+		req, _ := http.NewRequest("POST", setup.Server.URL+"/api/collections", bytes.NewBuffer(body))
+		req.Header.Set("Content-Type", "application/json")
+		req.Header.Set("Authorization", "Bearer "+setup.Token)
+		resp, err := http.DefaultClient.Do(req)
+		require.NoError(t, err)
 		require.Equal(t, http.StatusCreated, resp.StatusCode)
+		var created struct {
+			ID string `json:"id"`
+		}
+		require.NoError(t, json.NewDecoder(resp.Body).Decode(&created))
 		_ = resp.Body.Close()
+		collID := created.ID
+		t.Cleanup(func() {
+			del, _ := http.NewRequest("DELETE", setup.Server.URL+"/api/collections/"+collID, nil)
+			del.Header.Set("Authorization", "Bearer "+setup.Token)
+			_, _ = http.DefaultClient.Do(del)
+		})
 
 		addBody, _ := json.Marshal(map[string]interface{}{"book_ids": []string{uuid.NewString()}})
 		addReq, _ := http.NewRequest("POST", setup.Server.URL+"/api/collections/"+collID+"/books", bytes.NewBuffer(addBody))
@@ -238,53 +210,10 @@ func TestSmartCollections(t *testing.T) {
 		_ = removeResp.Body.Close()
 	})
 
-	t.Run("RemoveLastRuleRevertsToStatic", func(t *testing.T) {
+	t.Run("SystemCollectionsBlockedFromTheEditor", func(t *testing.T) {
-		smartCreateLibraryAndBooks(t, setup, adminID, map[string]string{
+		// System collections may never carry rules/trees — the editor
-			"Smart Back Book": "Science Fiction",
+		// route refuses them (item 3.4: the page IS the editor).
-		})
+		colls, err := setup.DB.GetCollectionsByUser(ownerSeedCtx(pgtype.UUID{Bytes: [16]byte(adminID), Valid: true}), pgtype.UUID{Bytes: [16]byte(adminID), Valid: true})
-		collID := smartCreateCollection(t, setup, setup.Token, "Test Smart Revert "+uuid.NewString()[:8], "smart")
-
-		resp := smartPostRule(t, setup, setup.Token, collID, "tags", "contains", "no-such-tag", nil)
-		require.Equal(t, http.StatusCreated, resp.StatusCode)
-		ruleID := smartDecodeRuleID(t, resp)
-
-		smartDetail := smartGetCollection(t, setup, setup.Token, collID)
-		assert.Equal(t, true, smartDetail["is_smart"])
-
-		delReq, _ := http.NewRequest("DELETE", setup.Server.URL+"/api/collections/"+collID+"/rules/"+ruleID, nil)
-		delReq.Header.Set("Authorization", "Bearer "+setup.Token)
-		delResp, err := http.DefaultClient.Do(delReq)
-		require.NoError(t, err)
-		require.Equal(t, http.StatusOK, delResp.StatusCode)
-		_ = delResp.Body.Close()
-
-		staticDetail := smartGetCollection(t, setup, setup.Token, collID)
-		assert.Equal(t, false, staticDetail["is_smart"], "removing the last rule reverts to static")
-
-		// Membership mutations work again on the reverted static.
-		addBody, _ := json.Marshal(map[string]interface{}{"book_ids": []string{uuid.NewString()}})
-		addReq, _ := http.NewRequest("POST", setup.Server.URL+"/api/collections/"+collID+"/books", bytes.NewBuffer(addBody))
-		addReq.Header.Set("Content-Type", "application/json")
-		addReq.Header.Set("Authorization", "Bearer "+setup.Token)
-		addResp, err := http.DefaultClient.Do(addReq)
-		require.NoError(t, err)
-		assert.Equal(t, http.StatusNoContent, addResp.StatusCode)
-		_ = addResp.Body.Close()
-	})
-
-	t.Run("RuleValidationAndSystemCollections", func(t *testing.T) {
-		collID := smartCreateCollection(t, setup, setup.Token, "Test Smart Valid "+uuid.NewString()[:8], "static")
-
-		resp := smartPostRule(t, setup, setup.Token, collID, "not_a_field", "equals", "x", nil)
-		require.Equal(t, http.StatusBadRequest, resp.StatusCode)
-		_ = resp.Body.Close()
-
-		resp = smartPostRule(t, setup, setup.Token, collID, "genre", "equals", "Science Fiction", nil)
-		require.Equal(t, http.StatusCreated, resp.StatusCode)
-		_ = resp.Body.Close()
-
-		// System collections may never carry rules.
-		colls, err := setup.DB.GetCollectionsByUser(context.Background(), pgtype.UUID{Bytes: [16]byte(adminID), Valid: true})
 		require.NoError(t, err)
 		var systemID string
 		for _, c := range colls {
@@ -295,20 +224,46 @@ func TestSmartCollections(t *testing.T) {
 		}
 		require.NotEmpty(t, systemID, "admin user should have system collections")
 
-		sysResp := smartPostRule(t, setup, setup.Token, systemID, "genre", "equals", "x", nil)
+		req, _ := http.NewRequest("GET", setup.Server.URL+"/collections/"+systemID+"/rules", nil)
-		require.Equal(t, http.StatusBadRequest, sysResp.StatusCode)
+		req.Header.Set("Authorization", "Bearer "+setup.Token)
-		_ = sysResp.Body.Close()
+		resp, err := http.DefaultClient.Do(req)
+		require.NoError(t, err)
+		defer func(Body io.ReadCloser) { _ = Body.Close() }(resp.Body)
+		assert.NotEqual(t, http.StatusOK, resp.StatusCode,
+			"system collections never render the smart editor")
 	})
 
 	t.Run("CollectionsListRespectsIncludeAutoWithLiveCounts", func(t *testing.T) {
 		_, _ = smartCreateLibraryAndBooks(t, setup, adminID, map[string]string{
 			"Smart List Book": "Science Fiction",
 		})
-		collID := smartCreateCollection(t, setup, setup.Token, "Test Smart List "+uuid.NewString()[:8], "smart")
+		tree := map[string]interface{}{
-
+			"scope": "books", "mode": "all",
-		resp := smartPostRule(t, setup, setup.Token, collID, "genre", "equals", "Science Fiction", nil)
+			"children": []map[string]interface{}{
-		require.Equal(t, http.StatusCreated, resp.StatusCode)
+				{"field": "genre", "op": "is", "values": []string{"Science Fiction"}},
-		_ = resp.Body.Close()
+			},
+		}
+		body, _ := json.Marshal(map[string]interface{}{
+			"name":        "Test Smart List " + uuid.NewString()[:8],
+			"filter_tree": tree,
+		})
+		req2, _ := http.NewRequest("POST", setup.Server.URL+"/api/collections", bytes.NewBuffer(body))
+		req2.Header.Set("Content-Type", "application/json")
+		req2.Header.Set("Authorization", "Bearer "+setup.Token)
+		resp2, err := http.DefaultClient.Do(req2)
+		require.NoError(t, err)
+		require.Equal(t, http.StatusCreated, resp2.StatusCode)
+		var created struct {
+			ID string `json:"id"`
+		}
+		require.NoError(t, json.NewDecoder(resp2.Body).Decode(&created))
+		_ = resp2.Body.Close()
+		collID := created.ID
+		t.Cleanup(func() {
+			del, _ := http.NewRequest("DELETE", setup.Server.URL+"/api/collections/"+collID, nil)
+			del.Header.Set("Authorization", "Bearer "+setup.Token)
+			_, _ = http.DefaultClient.Do(del)
+		})
 
 		// Default list (the Android app's read): smart collections hidden.
 		req, _ := http.NewRequest("GET", setup.Server.URL+"/api/collections", nil)

cmd/server/tests/contract_pins_test.go

@@ -459,6 +459,8 @@ var bookInfoKeys = []string{
 	"media_item_id", "title", "author", "cover_image_path",
 	"has_conflict", "format_group", "badge",
 	"viewer_percentage", "library_name",
+	// Episode-collections extras (2.3 widening, every surface emits):
+	"is_episode", "feed_title", "download_state", "hidden",
 }
 
 func collectionURL(setup *TestServerSetup, id string) string {
@@ -498,8 +500,9 @@ func TestContractPin_CollectionsJSON(t *testing.T) {
 	// payload; docs/contract_pins.md changelog).
 	envelopeKeys := []string{
 		"id", "user_id", "name", "description", "color", "icon",
-		"auto_assign_rules", "view_settings", "books", "book_count",
+		"view_settings", "books", "book_count",
 		"is_smart", "created_at", "up_next",
+		"filter_tree", // item-3.2 widening (docs/contract_pins.md)
 	}
 
 	t.Run("user collection with library_id (badged branch)", func(t *testing.T) {
@@ -562,11 +565,14 @@ func TestContractPin_CollectionsJSON(t *testing.T) {
 		assertArrayItemShape(t, "collection books[] (no library)", books, bookInfoKeys, bookInfoKeys)
 	})
 
-	t.Run("smart collection (rules branch)", func(t *testing.T) {
+	t.Run("smart collection (tree branch)", func(t *testing.T) {
 		resp := doReq(t, "POST", setup.Server.URL+"/api/collections", map[string]interface{}{
 			"name": fmt.Sprintf("Test Contract Smart %d", time.Now().UnixNano()),
-			"auto_assign_rules": []map[string]interface{}{
+			"filter_tree": map[string]interface{}{
-				{"id": "rule-1", "field": "genre", "operator": "equals", "value": "Science Fiction", "priority": 5},
+				"scope": "books", "mode": "any",
+				"children": []map[string]interface{}{
+					{"field": "genre", "op": "is", "values": []string{"Science Fiction"}},
+				},
 			},
 		}, setup.Token)
 		defer resp.Body.Close()
@@ -752,6 +758,19 @@ func TestContractPin_MediaListBothBranches(t *testing.T) {
 		"tags_search", "contributors_search", "narrators_search", "file_sha256",
 		"opf_identifier", "opf_uuid", "hash_confidence", "library_name",
 		"library_type_name_2",
+		// Podcast episode columns (2026-10-29 additive widening — plan
+		// podcast-integration.md tasks 1.1/1.10): columns on
+		// media_items, NULL for scan rows. See docs/contract_pins.md
+		// changelog.
+		"source", "source_feed_id", "episode_guid", "feed_position",
+		"season_number", "episode_number", "enclosure_url",
+		"declared_duration_seconds", "transcript_url", "chapters_url",
+		"hosts", "guests", "download_state", "download_error",
+		"download_attempts",
+		// §0 item 4 (2026-10-30 additive widening): the download pool's
+		// queue ticket — SEQUENCE-stamped at enqueue, NULL for scan rows
+		// and never-queued episodes. See docs/contract_pins.md changelog.
+		"download_queue_pos",
 	}
 	resolvedKeys := []string{
 		"id", "library_id", "title", "subtitle", "author", "authors", "isbn", "description",

cmd/server/tests/custom_section_builder_test.go

@@ -0,0 +1,157 @@
+package main
+
+import (
+	"encoding/json"
+	"fmt"
+	"net/http"
+	"strings"
+	"testing"
+	"time"
+
+	"github.com/google/uuid"
+	"github.com/jackc/pgx/v5/pgtype"
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+
+	"bookhoard/internal/database"
+)
+
+// The custom-section builder's save, end to end (2026-10-31 tree
+// migration). The page posts the legacy rules vocabulary — rules as a
+// JSON-encoded ARRAY inside a STRING (it has done so since the page was
+// written, 2026-02-19; the save never once bound successfully before
+// the migration) — plus match_type, hand-picked book ids, and the
+// dashboard pin. The server converts the rules to a filter tree at
+// write time, attaches the manual picks, and the dashboard section
+// renders stored membership + the tree's live matches, deduped.
+func TestCustomSectionBuilderSave(t *testing.T) {
+	setup := setupTestServer(t)
+	adminToken := setup.Token
+	regToken := setup.RegularToken
+	marker := fmt.Sprintf("csb%d", time.Now().UnixNano()%100000)
+
+	libID := createTestLibraryWithFolder(t, setup.Server, adminToken, "test CSB Library "+marker, true)
+	libUUID, err := uuid.Parse(libID)
+	require.NoError(t, err)
+
+	// Two items the title rule matches, one it doesn't. Alpha is ALSO
+	// hand-picked — the section must render it once, not twice.
+	makeItem := func(title string) uuid.UUID {
+		id, cerr := setup.DB.CreateMediaItem(seedCtx(), database.CreateMediaItemParams{
+			LibraryID: pgtype.UUID{Bytes: libUUID, Valid: true},
+			Title:     title,
+			Authors:   []string{"CSB Author"},
+			FilePath:  fmt.Sprintf("/tmp/csb/%s-%s.epub", marker, strings.ReplaceAll(title, " ", "-")),
+			FileSize:  pgtype.Int8{Int64: 128, Valid: true},
+			MimeType:  pgtype.Text{String: "application/epub+zip", Valid: true},
+		})
+		require.NoError(t, cerr)
+		out, perr := uuid.FromBytes(id.ID.Bytes[:])
+		require.NoError(t, perr)
+		return out
+	}
+	alpha := makeItem("CSBRULE Alpha")
+	_ = makeItem("CSBRULE Beta")
+	manual := makeItem("CSB Manual Pick")
+	require.NotZero(t, alpha)
+
+	// The EXACT payload shape the builder page posts.
+	payload := fmt.Sprintf(`{
+		"library_id": %q,
+		"name": "test CSB Section %s",
+		"icon": "📚",
+		"description": "built by the custom-section builder",
+		"show_on_dashboard": true,
+		"auto_assign_rules": %[3]s,
+		"manual_book_ids": [%q, %q],
+		"match_type": "all"
+	}`, libID, marker,
+		`"[{\"field\":\"title\",\"operator\":\"contains\",\"value\":\"CSBRULE\"}]"`,
+		alpha.String(), manual.String())
+
+	code, body := podcastAuthed(t, setup, regToken, "POST", "/api/collections", payload)
+	require.Equal(t, http.StatusCreated, code, "%s", string(body))
+	var coll struct {
+		ID string `json:"id"`
+	}
+	require.NoError(t, json.Unmarshal([]byte(body), &coll))
+	collID := coll.ID
+	require.NotEmpty(t, collID)
+	t.Cleanup(func() {
+		podcastAuthed(t, setup, regToken, "DELETE", "/api/collections/"+collID, "")
+	})
+
+	// The converted tree round-trips: root mode from match_type, the
+	// rule as a qf child, and the collection IS smart with no
+	// collection_type sent.
+	code, body = podcastAuthed(t, setup, regToken, "GET", "/api/collections/"+collID, "")
+	require.Equal(t, http.StatusOK, code)
+	var detail struct {
+		IsSmart    bool `json:"is_smart"`
+		FilterTree struct {
+			Mode     string `json:"mode"`
+			Children []struct {
+				Field  string   `json:"field"`
+				Op     string   `json:"op"`
+				Values []string `json:"values"`
+			} `json:"children"`
+		} `json:"filter_tree"`
+	}
+	require.NoError(t, json.Unmarshal([]byte(body), &detail))
+	assert.True(t, detail.IsSmart, "converted rules make the collection smart")
+	assert.Equal(t, "all", detail.FilterTree.Mode, "match_type seeds the tree's root mode")
+	require.Len(t, detail.FilterTree.Children, 1)
+	assert.Equal(t, "title", detail.FilterTree.Children[0].Field)
+	assert.Equal(t, []string{"CSBRULE"}, detail.FilterTree.Children[0].Values)
+
+	// The dashboard section renders stored membership + live matches,
+	// deduped: Alpha (rule + manual), Beta (rule), Manual (manual).
+	code, body = podcastAuthed(t, setup, regToken, "GET",
+		"/api/dashboard/sections?library_id="+libID, "")
+	require.Equal(t, http.StatusOK, code, "%s", string(body))
+	var dash struct {
+		Sections []struct {
+			Title        string `json:"title"`
+			CollectionID string `json:"collection_id"`
+			Items        []struct {
+				Title string `json:"title"`
+			} `json:"items"`
+		} `json:"sections"`
+	}
+	require.NoError(t, json.Unmarshal([]byte(body), &dash))
+	var section *struct {
+		Title        string `json:"title"`
+		CollectionID string `json:"collection_id"`
+		Items        []struct {
+			Title string `json:"title"`
+		} `json:"items"`
+	}
+	for i := range dash.Sections {
+		if dash.Sections[i].CollectionID == collID {
+			section = &dash.Sections[i]
+			break
+		}
+	}
+	require.NotNil(t, section, "the custom section renders on the dashboard: %s", string(body))
+	titles := map[string]int{}
+	for _, it := range section.Items {
+		titles[it.Title]++
+	}
+	assert.Equal(t, 1, titles["CSBRULE Alpha"], "rule-matched AND hand-picked renders exactly once")
+	assert.Equal(t, 1, titles["CSBRULE Beta"], "rule match rides the tree")
+	assert.Equal(t, 1, titles["CSB Manual Pick"], "manual picks are attached")
+	assert.Len(t, section.Items, 3, "nothing else leaks into the section")
+}
+
+// TestCustomSectionBuilderBadRule pins the adapter's rejection path: a
+// rule the engine cannot honor is a 400 that names the field, never a
+// stored 500 at render time.
+func TestCustomSectionBuilderBadRule(t *testing.T) {
+	setup := setupTestServer(t)
+	regToken := setup.RegularToken
+
+	code, body := podcastAuthed(t, setup, regToken, "POST", "/api/collections",
+		`{"name":"test CSB Bad `+fmt.Sprint(time.Now().UnixNano())+`","auto_assign_rules":"[{\"field\":\"not_a_field\",\"operator\":\"contains\",\"value\":\"x\"}]"}`)
+	assert.Equal(t, http.StatusBadRequest, code, "%s", string(body))
+	assert.Contains(t, string(body), "section rule")
+}

cmd/server/tests/filter_builder_test.go

@@ -295,6 +295,115 @@ func TestFilterComposedSemantics(t *testing.T) {
 	assert.Equal(t, []string{"Sem Tide"}, got, "published_year at_least")
 }
 
+// TestFilterScopesAndEpisodeFields pins item 3.1: the books scope
+// excludes podcasts (the leak the composed head used to have), the
+// episodes scope is episodes only, episode fields evaluate in SQL
+// (feed-in, downloaded, played, published-date-within), and episode
+// fields are 400s in books trees.
+func TestFilterScopesAndEpisodeFields(t *testing.T) {
+	setup := setupTestServer(t)
+	t.Cleanup(func() { _ = setup.Close() })
+
+	regUUID := getRegularUserID(t, setup.DB)
+	builderFixtureLib(t, setup, regUUID, "Scope")
+
+	// Podcasts library + a subscribed feed with three downloaded
+	// episodes: two fresh/unplayed, one played (~finished) — the
+	// played-rule fixture.
+	feedLib, _ := createPodcastsLibrary(t, setup, "test Scope Podcasts")
+	feed, err := setup.DB.CreatePodcastFeed(adminSeedCtx(t, setup), database.CreatePodcastFeedParams{
+		LibraryID: feedLib.ID,
+		FeedUrl:   "https://feed.invalid/scope-probe.xml",
+		Title:     "Scope Probe Feed",
+	})
+	require.NoError(t, err)
+	epFresh := upsertTestEpisode(t, setup, feed, "scope-fresh", "Scope Fresh Episode", 1)
+	epPlayed := upsertTestEpisode(t, setup, feed, "scope-played", "Scope Played Episode", 2)
+	upsertTestEpisode(t, setup, feed, "scope-pending", "Scope Pending Episode", 3)
+	subscribeRaw(t, setup, regUUID, feed.ID)
+	adminScopeExec(t, setup, adminCtxValue(t, setup), `
+		UPDATE media_items SET download_state = 'downloaded', file_size = 1000,
+		       date_published = CURRENT_DATE WHERE id IN ($1, $2)`, epFresh.ID, epPlayed.ID)
+	adminScopeExec(t, setup, adminCtxValue(t, setup), `
+		INSERT INTO reading_progress (media_item_id, user_id, percentage)
+		VALUES ($1, $2, 1.0) ON CONFLICT (media_item_id, user_id)
+		DO UPDATE SET percentage = 1.0`, epPlayed.ID, regUUID)
+
+	feedID := uuid.UUID(feed.ID.Bytes).String()
+
+	qfScoped := func(scope string, children ...map[string]interface{}) url.Values {
+		b, _ := json.Marshal(map[string]interface{}{"scope": scope, "mode": "all", "children": children})
+		return url.Values{"qf": {string(b)}}
+	}
+	const booksScope = "books"
+	const episodesScope = "episodes"
+
+	search := func(p url.Values) []string {
+		rows := p
+		status, body := qfSearch(t, setup, setup.RegularToken, rows)
+		// 404 is the endpoint's pinned no-results contract — an empty set.
+		require.Contains(t, []int{http.StatusOK, http.StatusNotFound}, status, "qf must parse: %v", rows)
+		return qfTitles(t, body)
+	}
+
+	t.Run("books_scope_excludes_podcasts", func(t *testing.T) {
+		// An exact-ish title hits BOTH a book fixture ("Scope Tide") and
+		// the episode fixture if the head leaks; books scope must return
+		// the book, never the episodes.
+		got := search(qfScoped(booksScope, qfRule("title", "is", "Scope Tide")))
+		assert.Equal(t, []string{"Scope Tide"}, got, "books scope returns the book")
+		got = search(qfScoped(booksScope, qfRule("title", "is", "Scope Fresh Episode")))
+		assert.Empty(t, got,
+			"books scope must never return podcast episodes (the composed-head leak is CLOSED)")
+	})
+
+	t.Run("episodes_scope_is_episodes_only", func(t *testing.T) {
+		got := search(qfScoped(episodesScope, qfRule("title", "is", "Scope Fresh Episode")))
+		assert.Equal(t, []string{"Scope Fresh Episode"}, got,
+			"episodes scope: only episodes — books drop")
+	})
+
+	t.Run("episode_rules_evaluate_in_sql", func(t *testing.T) {
+		got := search(qfScoped(episodesScope,
+			qfRule("feed", "in", feedID)))
+		assert.ElementsMatch(t, []string{"Scope Fresh Episode", "Scope Played Episode", "Scope Pending Episode"},
+			got, "feed-in matches every episode of the feed (download state is a separate rule)")
+
+		got = search(qfScoped(episodesScope,
+			qfRule("downloaded", "is_true")))
+		assert.ElementsMatch(t, []string{"Scope Fresh Episode", "Scope Played Episode"},
+			got, "downloaded:true drops the pending episode")
+
+		got = search(qfScoped(episodesScope,
+			qfRule("played", "is_false"),
+			qfRule("downloaded", "is_true")))
+		assert.ElementsMatch(t, []string{"Scope Fresh Episode"}, got,
+			"played:false + downloaded = the unplayed playable episode (viewer progress)")
+
+		got = search(qfScoped(episodesScope,
+			qfRule("played", "is_true")))
+		assert.ElementsMatch(t, []string{"Scope Played Episode"}, got,
+			"played:true matches the finished episode")
+
+		got = search(qfScoped(episodesScope,
+			qfRule("published_date", "within_days", "7"),
+			qfRule("downloaded", "is_true")))
+		assert.ElementsMatch(t, []string{"Scope Fresh Episode", "Scope Played Episode"},
+			got, "published-within includes the dated episodes")
+	})
+
+	t.Run("episode_fields_are_400_in_books_scope", func(t *testing.T) {
+		p := qfScoped(booksScope, qfRule("played", "is_true"))
+		req, _ := http.NewRequest("GET", setup.Server.URL+"/api/media-items/search?"+p.Encode(), nil)
+		req.Header.Set("Authorization", "Bearer "+setup.RegularToken)
+		resp, err := http.DefaultClient.Do(req)
+		require.NoError(t, err)
+		defer func() { _ = resp.Body.Close() }()
+		assert.Equal(t, http.StatusBadRequest, resp.StatusCode,
+			"episode-only fields are rejected in books trees at parse time")
+	})
+}
+
 func TestFilterUserRelativeFields(t *testing.T) {
 	setup := setupTestServer(t)
 	t.Cleanup(func() { _ = setup.Close() })
@@ -346,12 +455,17 @@ func TestFilterUserRelativeFields(t *testing.T) {
 		aID, wickID, aID, tideID)
 
 	var collID pgtype.UUID
-	err := setup.DBPool.QueryRow(context.Background(),
+	// Seed as the OWNER (collections RLS, 2026-10-31): the insert
+	// predicate wants actor == owner, and the item/cleanup writes want
+	// the same identity — the admin write-escape covers SYSTEM rows only.
+	err := database.NewScopedDB(setup.DBPool).QueryRow(ownerSeedCtx(aID),
 		`INSERT INTO collections (user_id, name) VALUES ($1, $2) RETURNING id`,
 		aID, fmt.Sprintf("Test Builder Coll %d", time.Now().UnixNano())).Scan(&collID)
 	require.NoError(t, err)
-	t.Cleanup(func() { _ = setup.DB.DeleteCollection(ctx, collID) })
+	t.Cleanup(func() {
-	seedExec(t, setup.DBPool,
+		scopeExecAs(t, setup, aID, `DELETE FROM collections WHERE id = $1`, collID)
+	})
+	scopeExecAs(t, setup, aID,
 		`INSERT INTO collection_items (collection_id, media_item_id) VALUES ($1,$2)`, collID, tideID)
 
 	titlesFor := func(token string, qf string) []string {
@@ -433,7 +547,6 @@ func TestFilterSuggestionsUnderRules(t *testing.T) {
 	libUUID, _ := builderFixtureLib(t, setup, userID, "Sug")
 
 	// Add a second distinct author+genre pair (trigram-disjoint).
-	ctx := context.Background()
 	_, err := setup.DB.CreateMediaItem(seedCtx(), database.CreateMediaItemParams{
 		LibraryID: pgtype.UUID{Bytes: libUUID, Valid: true},
 		Title:     "Sug Dune",
@@ -537,6 +650,16 @@ func TestFilterSuggestionsUnderRules(t *testing.T) {
 	}
 	assert.True(t, found, "the visible library is offered")
 
+	// ID LOOKUP (2026-11-01): typing a stored ID resolves its label —
+	// the builder's post-Apply reload path resolves stored ids to names
+	// through this same call.
+	status, results = suggest("library", libUUID.String(), "")
+	require.Equal(t, http.StatusOK, status)
+	require.NotEmpty(t, results, "the id lookup returns its row")
+	assert.Equal(t, libUUID.String(), results[0]["value"])
+	disp, _ := results[0]["display"].(string)
+	assert.Contains(t, disp, "Test Builder", "the id resolves to the library NAME")
+
 	// BROWSE the library field UNDER rules (the blank-typed click): the
 	// id-source query must not choke on the unused typed parameter and
 	// must list exactly the library holding the surviving books.
@@ -561,16 +684,21 @@ func TestFilterSuggestionsUnderRules(t *testing.T) {
 	// offered with the surviving count; zero-match containers never are.
 	userA := getRegularUserID(t, setup.DB)
 	var collID pgtype.UUID
-	err = setup.DBPool.QueryRow(context.Background(),
+	// Seed as the OWNER (collections RLS, 2026-10-31) — same identity
+	// rule as the first seeding site in this file.
+	ownerID := pgtype.UUID{Bytes: [16]byte(userA), Valid: true}
+	err = database.NewScopedDB(setup.DBPool).QueryRow(ownerSeedCtx(ownerID),
 		`INSERT INTO collections (user_id, name) VALUES ($1, $2) RETURNING id`,
-		pgtype.UUID{Bytes: [16]byte(userA), Valid: true},
+		ownerID,
 		fmt.Sprintf("Test Builder Sug Coll %d", time.Now().UnixNano())).Scan(&collID)
 	require.NoError(t, err)
-	t.Cleanup(func() { _ = setup.DB.DeleteCollection(ctx, collID) })
+	t.Cleanup(func() {
+		scopeExecAs(t, setup, ownerID, `DELETE FROM collections WHERE id = $1`, collID)
+	})
 	var tideID pgtype.UUID
 	err = setup.DBPool.QueryRow(context.Background(), `SELECT id FROM media_items WHERE title = $1`, "Sug Tide").Scan(&tideID)
 	require.NoError(t, err)
-	seedExec(t, setup.DBPool,
+	scopeExecAs(t, setup, ownerID,
 		`INSERT INTO collection_items (collection_id, media_item_id) VALUES ($1, $2)`, collID, tideID)
 
 	status, results = suggest("collection", " ", "")
@@ -775,3 +903,85 @@ func TestFilterFlatEquivalence(t *testing.T) {
 	require.Equal(t, statusFlat, statusTree, "same result status")
 	assert.ElementsMatch(t, qfTitles(t, flatRows), qfTitles(t, treeRows))
 }
+
+// ---- deep group nesting (2026-11-01 group ALL/ANY header toggle): ----
+// the UI can now produce 3+-level mixed-mode trees; pin the engine's
+// parenthesized evaluation for the user's target structure
+// GROUP(GROUP(Field and (Field or Field)) or GROUP(Field and Field and
+// (Field or Field)) and GROUP(Field)) — plus a suggestion request
+// carrying a nested-any constraint (the mode-aware menu under real
+// nesting) — must be 200, never the empty-group 400.
+
+func TestFilterDeepGroupNesting(t *testing.T) {
+	setup := setupTestServer(t)
+	t.Cleanup(func() { _ = setup.Close() })
+
+	userID := getRegularUserID(t, setup.DB)
+	libUUID, _ := builderFixtureLib(t, setup, userID, "Deep")
+
+	titles := func(qf string) []string {
+		p := url.Values{"library_id": {libUUID.String()}, "qf": {qf}}
+		status, rows := qfSearch(t, setup, setup.RegularToken, p)
+		require.Contains(t, []int{http.StatusOK, http.StatusNotFound}, status)
+		return qfTitles(t, rows)
+	}
+
+	// root ALL[
+	//   ANY[ ALL[fantasy, ANY[>=250p, abridged]] ,  ALL[Dar, >=500p] ],
+	//   author not_contains Quill
+	// ]
+	// => Tide (fantasy + 300p), Wick (Dar + 700p); Fen dies at the root
+	// AND (Quill); Stone fails branch A (history) and branch B (Voss).
+	deep := map[string]interface{}{
+		"mode": "all",
+		"children": []interface{}{
+			map[string]interface{}{
+				"mode": "any",
+				"children": []interface{}{
+					map[string]interface{}{
+						"mode": "all",
+						"children": []interface{}{
+							qfRule("genre", "is", "fantasy"),
+							map[string]interface{}{
+								"mode": "any",
+								"children": []interface{}{
+									qfRule("page_count", "at_least", "250"),
+									qfRule("abridged", "is_true"),
+								},
+							},
+						},
+					},
+					map[string]interface{}{
+						"mode": "all",
+						"children": []interface{}{
+							qfRule("author", "contains", "Dar"),
+							qfRule("page_count", "at_least", "500"),
+						},
+					},
+				},
+			},
+			qfRule("author", "not_contains", "Quill"),
+		},
+	}
+	deepJSON, merr := json.Marshal(deep)
+	require.NoError(t, merr)
+	got := titles(string(deepJSON))
+	assert.ElementsMatch(t, []string{"Deep Tide", "Deep Wick"}, got,
+		"3-level mixed-mode nesting evaluates parenthesized")
+
+	// The same shape riding the SUGGEST endpoint (the mode-aware menu's
+	// constraint) — nested any must be a 200, never the empty-group 400.
+	p := url.Values{
+		"library_id": {libUUID.String()},
+		"author":     {""},
+		"limit":      {"20"},
+		"qf":         {string(deepJSON)},
+	}
+	req, _ := http.NewRequest("GET", setup.Server.URL+"/api/media-items/search?"+p.Encode(), nil)
+	req.Header.Set("Authorization", "Bearer "+setup.RegularToken)
+	resp, err := http.DefaultClient.Do(req)
+	require.NoError(t, err)
+	defer func() { _ = resp.Body.Close() }()
+	assert.Equal(t, http.StatusOK, resp.StatusCode,
+		"a nested-any constraint rides the suggestion endpoint cleanly: %s", p.Encode())
+}

cmd/server/tests/playlist_collections_test.go

@@ -13,6 +13,8 @@ import (
 
 	"github.com/google/uuid"
 	"github.com/stretchr/testify/require"
+
+	"bookhoard/internal/database"
 )
 
 // A2c end-to-end: an ABS sidecar whose series array mixes a real series and
@@ -67,13 +69,14 @@ func TestPlaylistCollectionsEndToEnd(t *testing.T) {
 	require.NotEmpty(t, foundPath)
 
 	// (b) The playlist collection exists with provenance, dashboard pin,
-	// and the item at reading position 7.
+	// and the item at reading position 7. Owner-scoped read: collections
+	// RLS (2026-10-31) hides the row from the pool session's SeedIdentity.
 	adminUUIDStr := adminID.String()
 	var collID uuid.UUID
 	var collName, collDesc string
 	var showOnDashboard bool
 	var viewSettingsRaw []byte
-	require.NoError(t, setup.DBPool.QueryRow(context.Background(),
+	require.NoError(t, database.NewScopedDB(setup.DBPool).QueryRow(ownerSeedCtx(adminID),
 		`SELECT id, name, description, show_on_dashboard, view_settings
 		 FROM collections WHERE user_id = $1 AND name = 'The Tagged Saga (Reading Order)'`,
 		adminID).Scan(&collID, &collName, &collDesc, &showOnDashboard, &viewSettingsRaw))
@@ -85,7 +88,9 @@ func TestPlaylistCollectionsEndToEnd(t *testing.T) {
 
 	var position *int32
 	var itemCount int
-	require.NoError(t, setup.DBPool.QueryRow(context.Background(),
+	// Owner-scoped reads: collections RLS (2026-10-31) — items ride the
+	// parent's readability, so the pool session's SeedIdentity sees none.
+	require.NoError(t, database.NewScopedDB(setup.DBPool).QueryRow(ownerSeedCtx(adminID),
 		`SELECT ci.position, COUNT(*) OVER () FROM collection_items ci
 		 WHERE ci.collection_id = $1`, collID).Scan(&position, &itemCount))
 	require.Equal(t, 1, itemCount)
@@ -96,11 +101,11 @@ func TestPlaylistCollectionsEndToEnd(t *testing.T) {
 	// (c) Rescan: no duplicate collection, no duplicate membership.
 	scanTestFolder(t, setup, adminID, folder)
 	var collectionCount, membershipCount int
-	require.NoError(t, setup.DBPool.QueryRow(context.Background(),
+	require.NoError(t, database.NewScopedDB(setup.DBPool).QueryRow(ownerSeedCtx(adminID),
 		`SELECT COUNT(*) FROM collections WHERE user_id = $1 AND name = 'The Tagged Saga (Reading Order)'`,
 		adminID).Scan(&collectionCount))
 	require.Equal(t, 1, collectionCount)
-	require.NoError(t, setup.DBPool.QueryRow(context.Background(),
+	require.NoError(t, database.NewScopedDB(setup.DBPool).QueryRow(ownerSeedCtx(adminID),
 		`SELECT COUNT(*) FROM collection_items WHERE collection_id = $1`, collID).Scan(&membershipCount))
 	require.Equal(t, 1, membershipCount)
 
@@ -131,7 +136,9 @@ func TestPlaylistCollectionsEndToEnd(t *testing.T) {
 	// (e) hide_finished + Up Next + completion marker.
 	// Flip the collection's flag on, mark the single item finished for the
 	// admin, and the dashboard section becomes Complete with no items.
-	_, err = setup.DBPool.Exec(context.Background(),
+	// Owner-or-admin UPDATE policy: run as ADMIN (collections RLS,
+	// 2026-10-31).
+	adminScopeExec(t, setup, adminCtxValue(t, setup),
 		`UPDATE collections SET view_settings = view_settings || '{"hide_finished": true}'::jsonb WHERE id = $1`, collID)
 	require.NoError(t, err)
 	_, err = setup.DBPool.Exec(context.Background(),

cmd/server/tests/resume_rewind_test.go

@@ -0,0 +1,123 @@
+package main
+
+// Resume rewind (per-user server setting): seconds before the saved
+// position the audiobook/podcast player starts (0 = exact resume).
+// Contract: the profile endpoint owns reads/writes with a strict
+// 0/5/10 allow-list (the request shape is the keep_screen_awake string
+// precedent — the form select must be able to send an explicit Off),
+// and the chapters payload carries the value so every client (web +
+// Android, incl. the Auto service) gets it with its existing fetch.
+// All writes ride the fresh per-test regular user — the shared admin
+// account is never touched (AGENTS doctrine). Fixture book lives in a
+// test-named library that cascades away with the library delete.
+
+import (
+	"encoding/json"
+	"fmt"
+	"net/http"
+	"path/filepath"
+	"strings"
+	"testing"
+	"time"
+
+	"github.com/google/uuid"
+	"github.com/stretchr/testify/require"
+)
+
+func TestAudioResumeRewind(t *testing.T) {
+	setup := setupTestServer(t)
+	t.Cleanup(func() { _ = setup.Close() })
+
+	admin, _ := moveTestAdminID(t, setup)
+	libID, folder := audioTestLibrary(t, setup, fmt.Sprintf("Test Resume Rewind %d", time.Now().UnixNano()))
+	t.Cleanup(func() { deleteLibraries(t, setup, libID) })
+
+	// 664 Hz tone: not the shared helper's 440 or the speed test's 552,
+	// so the scanner's content dedup never skips this book.
+	writeTestTrackTone(t, filepath.Join(folder, "Rewind Book", "01 one.mp3"), 664)
+	scanTestFolder(t, setup, admin, folder)
+
+	var bookPath string
+	for _, item := range listLibraryAudioItems(t, setup, libID) {
+		if strings.Contains(filepath.ToSlash(item.FilePath), "Rewind Book") && bookPath == "" {
+			bookPath = item.FilePath
+		}
+	}
+	require.NotEmpty(t, bookPath, "Rewind Book item must exist")
+	book, found := findAudioItemByPath(t, setup, libID, bookPath)
+	require.True(t, found)
+	bookID := uuid.UUID(book.ID.Bytes)
+
+	client := &http.Client{}
+	token := setup.RegularToken
+	authed := func(method, url, body string) (*http.Response, func()) {
+		var req *http.Request
+		if body == "" {
+			req, _ = http.NewRequest(method, url, nil)
+		} else {
+			req, _ = http.NewRequest(method, url, strings.NewReader(body))
+			req.Header.Set("Content-Type", "application/json")
+		}
+		req.Header.Set("Authorization", "Bearer "+token)
+		resp, err := client.Do(req)
+		require.NoError(t, err)
+		return resp, func() { _ = resp.Body.Close() }
+	}
+	getPayload := func(itemID uuid.UUID) (rewind int) {
+		resp, closer := authed("GET", setup.Server.URL+"/api/media-items/"+itemID.String()+"/chapters", "")
+		defer closer()
+		require.Equal(t, http.StatusOK, resp.StatusCode)
+		var data struct {
+			ResumeRewindSeconds int `json:"resume_rewind_seconds"`
+		}
+		require.NoError(t, json.NewDecoder(resp.Body).Decode(&data))
+		return data.ResumeRewindSeconds
+	}
+	getProfileRewind := func() int {
+		resp, closer := authed("GET", setup.Server.URL+"/api/auth/profile", "")
+		defer closer()
+		require.Equal(t, http.StatusOK, resp.StatusCode)
+		var profile struct {
+			AudioResumeRewind int `json:"audio_resume_rewind_seconds"`
+		}
+		require.NoError(t, json.NewDecoder(resp.Body).Decode(&profile))
+		return profile.AudioResumeRewind
+	}
+
+	t.Run("defaults to off (exact resume)", func(t *testing.T) {
+		require.Equal(t, 0, getProfileRewind())
+		require.Equal(t, 0, getPayload(bookID))
+	})
+
+	t.Run("profile PUT selects 5 and rides the chapters payload", func(t *testing.T) {
+		resp, closer := authed("PUT", setup.Server.URL+"/api/auth/profile", `{"audio_resume_rewind_seconds":"5"}`)
+		defer closer()
+		require.Equal(t, http.StatusOK, resp.StatusCode)
+		require.Equal(t, 5, getProfileRewind())
+		require.Equal(t, 5, getPayload(bookID))
+	})
+
+	t.Run("non-allow-listed values are rejected and change nothing", func(t *testing.T) {
+		for _, body := range []string{`{"audio_resume_rewind_seconds":"7"}`, `{"audio_resume_rewind_seconds":"-5"}`, `{"audio_resume_rewind_seconds":"abc"}`} {
+			resp, closer := authed("PUT", setup.Server.URL+"/api/auth/profile", body)
+			require.Equal(t, http.StatusBadRequest, resp.StatusCode)
+			closer()
+		}
+		require.Equal(t, 5, getProfileRewind())
+	})
+
+	t.Run("explicit off resets to 0 everywhere", func(t *testing.T) {
+		resp, closer := authed("PUT", setup.Server.URL+"/api/auth/profile", `{"audio_resume_rewind_seconds":"0"}`)
+		defer closer()
+		require.Equal(t, http.StatusOK, resp.StatusCode)
+		require.Equal(t, 0, getProfileRewind())
+		require.Equal(t, 0, getPayload(bookID))
+	})
+
+	t.Run("JSON clients that omit the key leave it untouched", func(t *testing.T) {
+		resp, closer := authed("PUT", setup.Server.URL+"/api/auth/profile", `{"timezone":"UTC"}`)
+		defer closer()
+		require.Equal(t, http.StatusOK, resp.StatusCode)
+		require.Equal(t, 0, getProfileRewind())
+	})
+}

cmd/server/tests/scan_state_test.go

@@ -207,12 +207,13 @@ func TestScanManifestLifecycle(t *testing.T) {
 	require.NoError(t, os.WriteFile(opfPath, []byte(`{"title":"Manifesu Book","series":["Lifecycle Saga (Reading Order) #1"]}`), 0644))
 	scanTestFolder(t, setup, admin, folder)
 	var collCount int
-	require.NoError(t, setup.DBPool.QueryRow(context.Background(),
+	// Owner-scoped read: collections RLS (2026-10-31).
+	require.NoError(t, database.NewScopedDB(setup.DBPool).QueryRow(ownerSeedCtx(admin),
 		`SELECT COUNT(*) FROM collections WHERE name = 'Lifecycle Saga (Reading Order)' AND user_id = $1`, admin,
 	).Scan(&collCount))
 	assert.Equal(t, 1, collCount, "reading-order sidecar added to an existing book creates its collection")
 	scanTestFolder(t, setup, admin, folder)
-	require.NoError(t, setup.DBPool.QueryRow(context.Background(),
+	require.NoError(t, database.NewScopedDB(setup.DBPool).QueryRow(ownerSeedCtx(admin),
 		`SELECT COUNT(*) FROM collections WHERE name = 'Lifecycle Saga (Reading Order)' AND user_id = $1`, admin,
 	).Scan(&collCount))
 	assert.Equal(t, 1, collCount, "rescan is idempotent — no duplicate collection")

cmd/server/tests/test_helpers_test.go

@@ -329,6 +329,22 @@ func seedCtx() context.Context {
 	return rls.WithUser(context.Background(), rls.SeedIdentity)
 }
 
+// ownerSeedCtx returns a context carrying a fixture user's identity,
+// for seeding rows OWNED by that user. Collections RLS (2026-10-31):
+// the write predicates want actor == owner for non-system rows (the
+// admin write-escape covers SYSTEM collections only), and reads under
+// the pool session's SeedIdentity see nothing.
+func ownerSeedCtx(userID pgtype.UUID) context.Context {
+	return rls.WithUser(context.Background(), userID)
+}
+
+// scopeExecAs is adminScopeExec's generic form: raw SQL under ANY
+// declared identity. Use it (with the row's OWNER) for fixture writes
+// to non-system collections — see ownerSeedCtx.
+func scopeExecAs(t *testing.T, setup *TestServerSetup, userID pgtype.UUID, sql string, args ...any) {
+	adminScopeExec(t, setup, userID, sql, args...)
+}
+
 // seedExec runs a raw statement against the pool under the seed
 // identity. The bare pool bypasses ScopedDB (nothing reads the ctx for
 // the GUC), so governed tables need the identity declared manually
@@ -356,7 +372,12 @@ func uuidToPGType(u uuid.UUID) pgtype.UUID {
 
 // createDefaultCollectionsForUser creates the 4 default system collections for a user
 func createDefaultCollectionsForUser(t *testing.T, db *database.Queries, userID pgtype.UUID) {
-	ctx := context.Background()
+	// Owner-scoped seeding: the collections RLS policies (2026-10-31)
+	// need an acting identity, and the harness seeds before any JWT —
+	// the same bootstrap fix as CreateDefaultCollectionsForUser in
+	// internal/handlers/auth.go. Owner == actor, so SELECT sees the
+	// rows, the INSERT passes, and the heal-UPDATE passes.
+	ctx := rls.WithUser(context.Background(), userID)
 	// Canonical Title-case names — the same list the registration seed
 	// (CreateDefaultCollectionsForUser) and RestoreSystemCollection use.
 	// A stale kebab-case copy lived here once and duplicated every
@@ -596,10 +617,13 @@ func setupTestServer(t *testing.T) *TestServerSetup {
 	worker := services.NewWorker(3, connManager)
 	services.WorkerInstance = worker
 	jobsHandler := handlers.NewJobsHandler(queries, worker)
-	collectionHandler := handlers.NewCollectionHandler(queries, libraryService, connManager)
+	// One composed-engine service for search AND the collection
+	// subsystem (item 3.2 — mirrors main.go).
+	collectionSearchService := services.NewSearchService(queries, scopedDB)
+	collectionHandler := handlers.NewCollectionHandler(queries, libraryService, connManager, collectionSearchService)
 	filtersHandler := handlers.NewFiltersHandler(queries)
-	dashboardService := services.NewDashboardService(queries)
+	dashboardService := services.NewDashboardService(queries, collectionSearchService)
-	dashboardHandler := handlers.NewDashboardHandler(queries)
+	dashboardHandler := handlers.NewDashboardHandler(queries, collectionSearchService)
 	statisticsService := services.NewStatisticsService(queries)
 	statsHandler := handlers.NewStatsHandler(statisticsService)
 	seriesHandler := handlers.NewSeriesHandler(queries)
@@ -610,6 +634,14 @@ func setupTestServer(t *testing.T) *TestServerSetup {
 	conversionService := services.NewConversionService(queries, getCachePath())
 	opdsHandler := handlers.NewOPDSHandler(queries, libraryService, conversionService)
 
+	// Podcast subsystem (mirrors main.go wiring): fetcher service +
+	// download worker + handler. The poller is NOT started in tests
+	// (no background ticks against the shared dev DB); the worker loop
+	// only drains when a test queues work.
+	podcastService := services.NewPodcastService(queries)
+	podcastWorker := services.NewPodcastDownloadWorker(queries, podcastService)
+	podcastHandler := handlers.NewPodcastHandler(queries, podcastService, podcastWorker)
+
 	// Create Echo instance
 	e := echo.New()
 
@@ -672,6 +704,9 @@ func setupTestServer(t *testing.T) *TestServerSetup {
 		// hands the kobo/kepub CFI converter a typed-nil (kobo markup
 		// panic-recovered as 500s in TestProgressWeb_KoboIntegration).
 		LibraryService: libraryService,
+		PodcastHandler: podcastHandler,
+		PodcastService: podcastService,
+		PodcastWorker:  podcastWorker,
 	}
 
 	router.RegisterRoutes(routerConfig)
@@ -758,11 +793,15 @@ func setupTestServer(t *testing.T) *TestServerSetup {
 	// run's creates die on UNIQUE(user_id, name) — the "collections
 	// 500-panic" rot. System collections are skipped (permanent rows).
 	// Same convention as the library pass: names containing "test".
-	adminColls, cerr := queries.GetCollectionsByUser(ctx, pgtype.UUID{Bytes: adminUUID, Valid: true})
+	// Swept under the OWNER's identity: collections RLS (2026-10-31)
+	// makes unscoped reads return nothing, which would silently turn
+	// this cleanup into a no-op.
+	adminScope := rls.WithUser(ctx, pgtype.UUID{Bytes: adminUUID, Valid: true})
+	adminColls, cerr := queries.GetCollectionsByUser(adminScope, pgtype.UUID{Bytes: adminUUID, Valid: true})
 	if cerr == nil {
 		for _, coll := range adminColls {
 			if !coll.IsSystemCollection.Bool && strings.Contains(strings.ToLower(coll.Name), "test") {
-				queries.DeleteCollection(ctx, coll.ID)
+				queries.DeleteCollection(adminScope, coll.ID)
 			}
 		}
 	}
@@ -836,11 +875,13 @@ func setupTestServer(t *testing.T) *TestServerSetup {
 
 		// Same pass for test-created COLLECTIONS (the "collections
 		// 500-panic" rot: they lingered forever and poisoned later runs).
-		adminColls, cerr := queries.GetCollectionsByUser(ctx, pgtype.UUID{Bytes: adminUUID, Valid: true})
+		// Owner-scoped for the same RLS reason as the setup sweep.
+		adminScope := rls.WithUser(ctx, pgtype.UUID{Bytes: adminUUID, Valid: true})
+		adminColls, cerr := queries.GetCollectionsByUser(adminScope, pgtype.UUID{Bytes: adminUUID, Valid: true})
 		if cerr == nil {
 			for _, coll := range adminColls {
 				if !coll.IsSystemCollection.Bool && strings.Contains(strings.ToLower(coll.Name), "test") {
-					queries.DeleteCollection(ctx, coll.ID)
+					queries.DeleteCollection(adminScope, coll.ID)
 				}
 			}
 		}

database/schema/schema.sql

@@ -18,7 +18,8 @@ INSERT INTO library_types (name, description, allowed_extensions) VALUES
 ('ebooks', 'Ebook files including EPUB, PDF, MOBI, etc.', ARRAY['.epub', '.pdf', '.mobi', '.azw', '.azw3', '.txt', '.rtf', '.doc', '.docx', '.lit', '.fb2', '.pdb']),
 ('comics', 'Comic book archives and image formats', ARRAY['.cbz', '.cbr', '.cb7', '.cbt', '.epub', '.pdf']),
 ('manga', 'Manga files including archives and image folders', ARRAY['.cbz', '.cbr', '.epub', '.pdf', '.png', '.jpg', '.jpeg', '.gif', '.bmp', '.webp', '.avif', '.tiff', '.tif']),
-('audiobooks', 'Audiobook files; m4b first, breadth formats registered from day one', ARRAY['.m4b', '.m4a', '.mp3', '.flac'])
+('audiobooks', 'Audiobook files; m4b first, breadth formats registered from day one', ARRAY['.m4b', '.m4a', '.mp3', '.flac']),
+('podcasts', 'RSS podcast/feed libraries: feeds subscribe-and-download; the scanner never walks these folders', ARRAY[]::TEXT[])
 ON CONFLICT (name) DO NOTHING;
 
 -- Create users table
@@ -47,6 +48,14 @@ ALTER TABLE users ADD COLUMN IF NOT EXISTS default_audio_speed REAL NOT NULL DEF
 -- mid-session. On by default; the profile page's On/Off dropdown owns it.
 ALTER TABLE users ADD COLUMN IF NOT EXISTS keep_screen_awake BOOLEAN NOT NULL DEFAULT TRUE;
 
+-- Audio resume rewind: seconds before the saved position the
+-- audiobook/podcast player starts (on open-resume AND again on every
+-- genuine unpause — the re-orientation gap, never stacked). Strict
+-- allow-list 0/5/10 enforced handler-side; 0 = exact resume. The
+-- profile page's dropdown owns it; the chapters payload carries it to
+-- every client (web + Android, incl. the car path).
+ALTER TABLE users ADD COLUMN IF NOT EXISTS audio_resume_rewind_seconds INT NOT NULL DEFAULT 0;
+
 -- Create system_settings table
 CREATE TABLE IF NOT EXISTS system_settings (
     id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
@@ -99,7 +108,11 @@ INSERT INTO system_settings (setting_key, setting_value, description, setting_ty
 ('sync_queue_interval_seconds', '5', 'How often the sync queue flushes', 'int', '1', '3600', TRUE, 'sync'),
 ('sync_queue_batch_size', '50', 'Maximum items processed per sync queue flush', 'int', '1', '10000', TRUE, 'sync'),
 ('worker_pool_size', '3', 'Number of background worker goroutines', 'int', '1', '100', TRUE, 'performance'),
-('worker_queue_cap', '100', 'Background worker job queue capacity', 'int', '1', '10000', TRUE, 'performance')
+('worker_queue_cap', '100', 'Background worker job queue capacity', 'int', '1', '10000', TRUE, 'performance'),
+-- podcasts
+('podcast_poll_interval_seconds', '3600', 'How often to poll podcast feeds for new episodes (seconds; complete feeds back off)', 'int', '300', '604800', FALSE, 'podcasts'),
+('podcast_download_concurrency', '4', 'Podcast episode downloads in flight server-wide (the connection is the limit; fair-share spreads them across feeds)', 'int', '1', '16', FALSE, 'podcasts'),
+('podcast_download_speed_mbps', '0', 'Shared bandwidth cap for podcast downloads in megabits/second (0 = unlimited)', 'float', '0', '10000', FALSE, 'podcasts')
 ON CONFLICT (setting_key) DO NOTHING;
 
 -- Create refresh_tokens table
@@ -1336,7 +1349,9 @@ CREATE TABLE IF NOT EXISTS collections (
     description TEXT,
     color VARCHAR(7),
     icon VARCHAR(50),
-    auto_assign_rules JSONB,
+    -- The unified filter tree (plan item 3.2): when present the qf
+    -- engine drives membership; see FilterTree. NULL = static.
+    filter_tree JSONB,
     view_settings JSONB,
     show_on_dashboard BOOLEAN DEFAULT false,
     query_type TEXT DEFAULT 'filter',
@@ -1352,12 +1367,23 @@ CREATE TABLE IF NOT EXISTS collections (
 -- the row id (and everything referencing it) never churn.
 ALTER TABLE collections ADD COLUMN IF NOT EXISTS deleted_at TIMESTAMPTZ;
 
+-- Legacy rules-array columns RETIRED (2026-10-31): the unified filter
+-- tree is the only smart-membership storage. No production row ever
+-- carried a rules array or a non-default match mode (verified), so the
+-- columns drop clean; the DROPs are the idempotent heal for databases
+-- created before this change.
+ALTER TABLE collections DROP COLUMN IF EXISTS auto_assign_rules;
+ALTER TABLE collections DROP COLUMN IF EXISTS rule_match_mode;
+
+-- Smart collections in the UNIFIED filter engine (plan item 3.2): the
+-- qf tree (same DSL as saved filters, See FilterTree). NULL = static.
+ALTER TABLE collections ADD COLUMN IF NOT EXISTS filter_tree JSONB;
+
 CREATE INDEX IF NOT EXISTS idx_collections_user_id ON collections(user_id);
 
 -- Index for dashboard queries
 CREATE INDEX IF NOT EXISTS idx_collections_dashboard ON collections(user_id, show_on_dashboard, priority)
     WHERE show_on_dashboard = true;
-
 -- Create user_dashboard_preferences table
 CREATE TABLE IF NOT EXISTS user_dashboard_preferences (
     id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
@@ -1623,12 +1649,19 @@ CREATE INDEX IF NOT EXISTS idx_dictionary_word ON dictionary_cache(word);
 CREATE TABLE IF NOT EXISTS reader_settings (
   id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
   user_id UUID NOT NULL REFERENCES users(id) ON DELETE CASCADE,
-  setting_key VARCHAR(50) NOT NULL,
+  setting_key VARCHAR(120) NOT NULL,
   setting_value JSONB NOT NULL,
   updated_at TIMESTAMPTZ DEFAULT NOW(),
   UNIQUE(user_id, setting_key)
 );
 
+-- Key-width headroom (podcast plan §10 R5, item 2.1): the composite
+-- `podcast_order:<uuid>` key sits at exactly 50 chars, and future
+-- per-feed/per-collection preference keys must not be forced to
+-- payload-stuff by a column width. Widening a varchar is metadata-only
+-- (no table rewrite); idempotent on every boot.
+ALTER TABLE reader_settings ALTER COLUMN setting_key TYPE VARCHAR(120);
+
 CREATE INDEX IF NOT EXISTS idx_reader_settings_user ON reader_settings(user_id);
 
 -- PDF bookmarks (custom user bookmarks)
@@ -1878,6 +1911,137 @@ CREATE INDEX IF NOT EXISTS idx_hash_conflicts_status ON hash_conflicts(status);
 -- Widen for databases created before the resolution format settled (no-op otherwise)
 ALTER TABLE hash_conflicts ALTER COLUMN resolution TYPE VARCHAR(50);
 
+-- ============================================================
+-- PODCASTS / RSS FEEDS
+-- Plan: ~/.opencode/plan/podcast-integration.md (§6).
+-- Feeds are per-library entities (UNIQUE(library_id, feed_url)); users
+-- subscribe personally (podcast_subscriptions). Episodes are
+-- media_items rows with source='podcast_episode' — per-episode
+-- progress/stats for free — grouped by source_feed_id. Visibility is
+-- RLS-enforced (see the policy block at the end of this file): an
+-- episode/feed row is SELECT-visible only to subscribers of the feed
+-- and to admins. The fetcher/download worker is the SOLE writer of
+-- feed/episode metadata; users own only their subscription row and the
+-- reader_settings order override. ALL DDL IDEMPOTENT: schema.sql
+-- re-executes on EVERY boot.
+-- ============================================================
+
+-- First per-library boolean knob (plain column, not KV): when false,
+-- only admins may add feeds to the library (members keep subscribe
+-- rights to existing feeds).
+ALTER TABLE libraries ADD COLUMN IF NOT EXISTS allow_member_feeds BOOLEAN NOT NULL DEFAULT TRUE;
+
+-- Folder purpose: 'scan' (default; the FS walker's root) vs 'podcast'
+-- (a podcasts library's download root — scanner skips it wholesale,
+-- belt-and-braces on top of the library-type skip).
+ALTER TABLE library_folders ADD COLUMN IF NOT EXISTS purpose VARCHAR(20) NOT NULL DEFAULT 'scan';
+
+CREATE TABLE IF NOT EXISTS podcast_feeds (
+    id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
+    library_id UUID NOT NULL REFERENCES libraries(id) ON DELETE CASCADE,
+    feed_url TEXT NOT NULL,
+    -- podcast:guid — the feed's canonical identity; dedupes across URL
+    -- migrations (fetcher honors itunes:new-feed-url + redirects).
+    podcast_guid TEXT,
+    title VARCHAR(500) NOT NULL,
+    description TEXT,
+    -- Channel itunes:author (semantically the host) and itunes:owner.
+    host VARCHAR(500),
+    publisher VARCHAR(500),
+    owner_name VARCHAR(500),
+    owner_email VARCHAR(500),
+    language VARCHAR(50),
+    tags TEXT[] DEFAULT '{}',
+    explicit BOOLEAN NOT NULL DEFAULT FALSE,
+    artwork_path VARCHAR(500),
+    -- itunes:type: 'serial' | 'episodic'
+    feed_type VARCHAR(20),
+    -- podcast:medium: 'audiobook' | 'podcast' | 'music' | ...
+    medium VARCHAR(20),
+    -- itunes:complete: poller backs off once the publisher ends the feed.
+    is_complete BOOLEAN NOT NULL DEFAULT FALSE,
+    -- Conditional-GET validators (opaque; echoed verbatim).
+    etag VARCHAR(500),
+    last_modified VARCHAR(500),
+    last_fetched_at TIMESTAMPTZ,
+    last_build_date TIMESTAMPTZ,
+    created_by_admin_id UUID REFERENCES users(id) ON DELETE SET NULL,
+    created_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
+    updated_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
+    UNIQUE(library_id, feed_url)
+);
+
+CREATE INDEX IF NOT EXISTS idx_podcast_feeds_library ON podcast_feeds(library_id);
+CREATE INDEX IF NOT EXISTS idx_podcast_feeds_guid ON podcast_feeds(podcast_guid);
+
+-- User-owned subscription row: the ONLY podcast state a user writes.
+-- display_name NULL renders as the feed title. auto_download:
+-- 'off' | 'all' | 'newest' (+ auto_download_count for newest N).
+CREATE TABLE IF NOT EXISTS podcast_subscriptions (
+    id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
+    user_id UUID NOT NULL REFERENCES users(id) ON DELETE CASCADE,
+    feed_id UUID NOT NULL REFERENCES podcast_feeds(id) ON DELETE CASCADE,
+    display_name VARCHAR(500),
+    auto_download VARCHAR(20) NOT NULL DEFAULT 'off',
+    auto_download_count INT,
+    created_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
+    UNIQUE(user_id, feed_id)
+);
+
+CREATE INDEX IF NOT EXISTS idx_podcast_subs_user ON podcast_subscriptions(user_id);
+CREATE INDEX IF NOT EXISTS idx_podcast_subs_feed ON podcast_subscriptions(feed_id);
+
+-- Episode columns (NULL for scan-sourced items). See the plan §5 for
+-- the full RSS/iTunes/Podcasting-2.0 field mapping — existing columns
+-- (title, subtitle, description, authors, explicit, date_published,
+-- cover_image_path, file_size, mime_type, language, genre/tags) carry
+-- the shared fields; only genuinely podcast-only storage lives here.
+ALTER TABLE media_items
+    ADD COLUMN IF NOT EXISTS source VARCHAR(20) NOT NULL DEFAULT 'scan',
+    ADD COLUMN IF NOT EXISTS source_feed_id UUID REFERENCES podcast_feeds(id) ON DELETE CASCADE,
+    -- Stable reorder-proof addressing key (invariant 3): progress
+    -- anchors on guid, never ordinal.
+    ADD COLUMN IF NOT EXISTS episode_guid TEXT,
+    -- Native order as of the latest fetch (renumbered per fetch).
+    ADD COLUMN IF NOT EXISTS feed_position INTEGER,
+    ADD COLUMN IF NOT EXISTS season_number INTEGER,
+    ADD COLUMN IF NOT EXISTS episode_number INTEGER,
+    ADD COLUMN IF NOT EXISTS enclosure_url TEXT,
+    -- itunes:duration claim (often HH:MM:SS or prose); display until
+    -- ffprobe reconciles after download.
+    ADD COLUMN IF NOT EXISTS declared_duration_seconds INTEGER,
+    ADD COLUMN IF NOT EXISTS transcript_url TEXT,
+    ADD COLUMN IF NOT EXISTS chapters_url TEXT,
+    ADD COLUMN IF NOT EXISTS hosts TEXT[] DEFAULT '{}',
+    ADD COLUMN IF NOT EXISTS guests TEXT[] DEFAULT '{}',
+    -- Download state machine: pending | queued | downloading |
+    -- downloaded | failed (+ reason text + attempt count). The episode
+    -- list renders this inline (D8/D19) — never Processing Issues.
+    ADD COLUMN IF NOT EXISTS download_state VARCHAR(20) NOT NULL DEFAULT 'pending',
+    ADD COLUMN IF NOT EXISTS download_error TEXT,
+    ADD COLUMN IF NOT EXISTS download_attempts INT NOT NULL DEFAULT 0,
+    -- Monotonic queue ticket (Postgres SEQUENCE nextval at enqueue
+    -- time): the download pool claims in ticket order, so the enqueue
+    -- loop's ordering IS the download order (effective listening order
+    -- — §0 items 4/5). NULL for rows never queued.
+    ADD COLUMN IF NOT EXISTS download_queue_pos BIGINT;
+
+-- Queue ticket source: atomic across concurrent enqueues (per-feed
+-- locks serialize within a feed; the sequence covers across feeds).
+CREATE SEQUENCE IF NOT EXISTS podcast_queue_seq;
+
+-- Identity + ordering + worker-worklist indexes (all partial: scan
+-- rows never touch these columns).
+CREATE UNIQUE INDEX IF NOT EXISTS idx_media_items_feed_guid
+    ON media_items(source_feed_id, episode_guid) WHERE source_feed_id IS NOT NULL;
+CREATE INDEX IF NOT EXISTS idx_media_items_feed_pos
+    ON media_items(source_feed_id, feed_position) WHERE source_feed_id IS NOT NULL;
+CREATE INDEX IF NOT EXISTS idx_media_items_feed_undownloaded
+    ON media_items(source_feed_id) WHERE source_feed_id IS NOT NULL AND download_state <> 'downloaded';
+CREATE INDEX IF NOT EXISTS idx_media_items_queue
+    ON media_items(download_queue_pos) WHERE source = 'podcast_episode' AND download_state = 'queued';
+
+
 -- ============================================================
 -- RLS: PER-USER LIBRARY VISIBILITY (ONE SOURCE OF TRUTH)
 -- Plan: ~/.opencode/plan/server-rls-library-visibility.md REV 2.
@@ -1912,6 +2076,13 @@ ALTER TABLE media_items FORCE ROW LEVEL SECURITY;
 -- semantics — row absent -> visible; row true -> visible; row false
 -- -> hidden. No new index needed: library_visibility's
 -- UNIQUE(user_id, library_id) constraint already indexes the join.
+-- PODCAST EPISODES (plan D20): source='podcast_episode' rows carry the
+-- ADDITIONAL subscription predicate — visible only to subscribers of
+-- source_feed_id, or to admins (admins manage feeds regardless of
+-- subscription; the escape also satisfies background jobs, which
+-- declare a first-admin identity). Scan rows pass unchanged via the
+-- IS DISTINCT FROM guard — the predicate costs them one NULL compare.
+-- Unset identity FAILS LOUD, never fails open (see the block header).
 CREATE POLICY p_media_items_visible ON media_items
   FOR SELECT
   USING (
@@ -1922,6 +2093,19 @@ CREATE POLICY p_media_items_visible ON media_items
         AND lv.library_id = media_items.library_id
         AND lv.is_visible = false
     )
+    AND (
+      media_items.source IS DISTINCT FROM 'podcast_episode'
+      OR EXISTS (
+        SELECT 1 FROM podcast_subscriptions s
+        WHERE s.user_id = current_setting('app.user_id')::uuid
+          AND s.feed_id = media_items.source_feed_id
+      )
+      OR EXISTS (
+        SELECT 1 FROM users u
+        WHERE u.id = current_setting('app.user_id')::uuid
+          AND u.role = 'admin'
+      )
+    )
   );
 
 -- WRITE policies: PERMISSIVE STAND-INS for the deferred scoped write
@@ -1942,7 +2126,173 @@ CREATE POLICY p_media_items_update ON media_items
 CREATE POLICY p_media_items_delete ON media_items
   FOR DELETE USING (true);
 
--- Other tables: NONE. collection_items reaches media only via INNER
+-- Other tables: collection_items reaches media only via INNER
 -- JOIN media_items (the join drops filtered rows). collections
 -- metadata (book_count) is a separate counts question — plan A13.
+
+-- PODCAST FEEDS (plan D20): governed the same way. SELECT-visible to
+-- subscribers of the feed (their subscription IS the grant — this is
+-- the predicate's lookup table, podcast_subscriptions itself stays
+-- UNGOVERNED) plus the library-visibility join (a hidden library
+-- hides its feeds too), and to admins. WRITE policies are permissive
+-- stand-ins for the same reason as media_items above: mandatory at
+-- ENABLE time, and the fetcher/worker (first-admin identity) plus the
+-- add-feed upsert (ON CONFLICT DO UPDATE needs no separate SELECT)
+-- cover every legitimate writer.
+DROP POLICY IF EXISTS p_podcast_feeds_visible ON podcast_feeds;
+DROP POLICY IF EXISTS p_podcast_feeds_insert ON podcast_feeds;
+DROP POLICY IF EXISTS p_podcast_feeds_update ON podcast_feeds;
+DROP POLICY IF EXISTS p_podcast_feeds_delete ON podcast_feeds;
+
+ALTER TABLE podcast_feeds ENABLE ROW LEVEL SECURITY;
+ALTER TABLE podcast_feeds FORCE ROW LEVEL SECURITY;
+
+CREATE POLICY p_podcast_feeds_visible ON podcast_feeds
+  FOR SELECT
+  USING (
+    current_setting('app.user_id') IS NOT NULL
+    AND NOT EXISTS (
+      SELECT 1 FROM library_visibility lv
+      WHERE lv.user_id = current_setting('app.user_id')::uuid
+        AND lv.library_id = podcast_feeds.library_id
+        AND lv.is_visible = false
+    )
+    AND (
+      EXISTS (
+        SELECT 1 FROM podcast_subscriptions s
+        WHERE s.user_id = current_setting('app.user_id')::uuid
+          AND s.feed_id = podcast_feeds.id
+      )
+      OR EXISTS (
+        SELECT 1 FROM users u
+        WHERE u.id = current_setting('app.user_id')::uuid
+          AND u.role = 'admin'
+      )
+    )
+  );
+
+CREATE POLICY p_podcast_feeds_insert ON podcast_feeds
+  FOR INSERT WITH CHECK (true);
+CREATE POLICY p_podcast_feeds_update ON podcast_feeds
+  FOR UPDATE USING (true) WITH CHECK (true);
+CREATE POLICY p_podcast_feeds_delete ON podcast_feeds
+  FOR DELETE USING (true);
+
+-- podcast_subscriptions: UNGOVERNED, deliberately — it IS the read
+-- policy's lookup table, and a user must always see their own
+-- subscription rows (subscribe -> visibility is a single INSERT).
 -- ============================================================
+-- ── collections + collection_items (2026-10-31): ownership at the DB ──
+-- layer, same doctrine as media_items/podcast_feeds above. The
+-- predicates live in FUNCTIONS (not inline) so a future sharing model
+-- edits exactly one place: add the share-table EXISTS branch inside
+-- app_can_read_collection / app_can_write_collection. The functions are
+-- PURE over their arguments — they never query collections — so the
+-- policies cannot recurse, FORCE RLS safe. Reads are authenticated-
+-- allow (household-readable: any signed-in identity, SeedIdentity
+-- harness included); WRITES are owner-or-admin — that is the
+-- enforcement of record for Update/Delete/AddBooks, which have no
+-- app-level gate.
+-- ===========================================================
+
+CREATE OR REPLACE FUNCTION app_is_admin() RETURNS BOOLEAN
+LANGUAGE sql STABLE SECURITY DEFINER SET search_path = pg_catalog, public AS $$
+  SELECT EXISTS (
+    SELECT 1 FROM users u
+    WHERE u.id = NULLIF(current_setting('app.user_id', true), '')::uuid
+      AND u.role = 'admin'
+  );
+$$;
+
+CREATE OR REPLACE FUNCTION app_can_read_collection(
+  p_collection_id UUID, p_owner UUID
+) RETURNS BOOLEAN
+LANGUAGE sql STABLE SECURITY DEFINER SET search_path = pg_catalog, public AS $$
+  SELECT NULLIF(current_setting('app.user_id', true), '') IS NOT NULL
+     AND (
+       p_owner = current_setting('app.user_id', true)::uuid
+       OR app_is_admin()
+       -- FUTURE SHARING slot: OR EXISTS (
+       --   SELECT 1 FROM collection_shares cs
+       --   WHERE cs.collection_id = p_collection_id
+       --     AND cs.shared_with_user_id = current_setting('app.user_id', true)::uuid)
+     );
+$$;
+
+CREATE OR REPLACE FUNCTION app_can_write_collection(
+  p_collection_id UUID, p_owner UUID, p_system BOOLEAN
+) RETURNS BOOLEAN
+LANGUAGE sql STABLE SECURITY DEFINER SET search_path = pg_catalog, public AS $$
+  SELECT NULLIF(current_setting('app.user_id', true), '') IS NOT NULL
+     AND (
+       p_owner = current_setting('app.user_id', true)::uuid
+       OR (COALESCE(p_system, false) AND app_is_admin())
+       -- FUTURE SHARING slot: OR EXISTS (
+       --   SELECT 1 FROM collection_shares cs
+       --   WHERE cs.collection_id = p_collection_id
+       --     AND cs.shared_with_user_id = current_setting('app.user_id', true)::uuid
+       --     AND cs.permission IN ('edit'))
+     );
+$$;
+
+ALTER TABLE collections ENABLE ROW LEVEL SECURITY;
+ALTER TABLE collections FORCE ROW LEVEL SECURITY;
+
+DROP POLICY IF EXISTS p_collections_select ON collections;
+CREATE POLICY p_collections_select ON collections
+  FOR SELECT USING (app_can_read_collection(id, user_id));
+DROP POLICY IF EXISTS p_collections_insert ON collections;
+CREATE POLICY p_collections_insert ON collections
+  FOR INSERT WITH CHECK (
+    user_id = NULLIF(current_setting('app.user_id', true), '')::uuid
+    OR app_is_admin()
+  );
+DROP POLICY IF EXISTS p_collections_update ON collections;
+CREATE POLICY p_collections_update ON collections
+  FOR UPDATE
+  USING (app_can_write_collection(id, user_id, is_system_collection))
+  WITH CHECK (app_can_write_collection(id, user_id, is_system_collection));
+DROP POLICY IF EXISTS p_collections_delete ON collections;
+CREATE POLICY p_collections_delete ON collections
+  FOR DELETE USING (app_can_write_collection(id, user_id, is_system_collection));
+
+ALTER TABLE collection_items ENABLE ROW LEVEL SECURITY;
+ALTER TABLE collection_items FORCE ROW LEVEL SECURITY;
+
+-- Items ride their PARENT's accessibility: the collections SELECT
+-- policy auto-applies inside these EXISTS subqueries.
+DROP POLICY IF EXISTS p_collection_items_select ON collection_items;
+CREATE POLICY p_collection_items_select ON collection_items
+  FOR SELECT USING (EXISTS (
+    SELECT 1 FROM collections c
+    WHERE c.id = collection_items.collection_id
+  ));
+DROP POLICY IF EXISTS p_collection_items_insert ON collection_items;
+CREATE POLICY p_collection_items_insert ON collection_items
+  FOR INSERT WITH CHECK (EXISTS (
+    SELECT 1 FROM collections c
+    WHERE c.id = collection_items.collection_id
+      AND app_can_write_collection(c.id, c.user_id, c.is_system_collection)
+  ));
+DROP POLICY IF EXISTS p_collection_items_update ON collection_items;
+CREATE POLICY p_collection_items_update ON collection_items
+  FOR UPDATE
+  USING (EXISTS (
+    SELECT 1 FROM collections c
+    WHERE c.id = collection_items.collection_id
+      AND app_can_write_collection(c.id, c.user_id, c.is_system_collection)
+  ))
+  WITH CHECK (EXISTS (
+    SELECT 1 FROM collections c
+    WHERE c.id = collection_items.collection_id
+      AND app_can_write_collection(c.id, c.user_id, c.is_system_collection)
+  ));
+DROP POLICY IF EXISTS p_collection_items_delete ON collection_items;
+CREATE POLICY p_collection_items_delete ON collection_items
+  FOR DELETE USING (EXISTS (
+    SELECT 1 FROM collections c
+    WHERE c.id = collection_items.collection_id
+      AND app_can_write_collection(c.id, c.user_id, c.is_system_collection)
+  ));
+
+

docker-compose.dev.yml

@@ -12,6 +12,8 @@ services:
     build:
       context: .
       dockerfile: ./Dockerfile
+      args:
+        GIT_COMMIT: ${GIT_COMMIT:-none}
 
   # Integration Tests - runs against containerized app and db (dev only)
   tests:

docs/contract_pins.md

@@ -443,3 +443,297 @@ Both answers are gated: 403 (Go-level explicit-library checks remain on
 listing endpoints) and 404 (row-level invisibility — no existence leak).
 Pins asserting 403 on item-scoped content endpoints widen to accept
 {403, 404}. Listing/scoped endpoints keep the strict 403 contract.
+
+## 2026-10-29 — Podcast episode columns join media_items (additive widening)
+
+The podcast integration (plan podcast-integration.md tasks 1.1/1.10) adds
+fifteen nullable/no-null-default columns to `media_items` — they surface in
+raw sqlc rows ONLY (`library-scoped list emits raw sqlc rows`), never in the
+resolved unscoped map:
+
+`source` ('scan' default), `source_feed_id`, `episode_guid`, `feed_position`,
+`season_number`, `episode_number`, `enclosure_url`,
+`declared_duration_seconds`, `transcript_url`, `chapters_url`, `hosts`,
+`guests`, `download_state`, `download_error`, `download_attempts`.
+
+Scan-sourced rows carry the inert defaults ('scan' / NULLs / 'pending');
+raw-row consumers that filter `source='scan'` (scanner paths) are unaffected.
+Nothing removed, nothing renamed.
+
+## 2026-10-30 — download_queue_pos joins media_items (additive widening)
+
+§0 item 4 of podcast-integration.md: the download pool's monotonic
+queue ticket. `download_queue_pos BIGINT` (NULL for scan rows and
+never-queued episodes) is stamped from the `podcast_queue_seq` SEQUENCE
+at enqueue time so claim order = enqueue order (the effective listening
+order). Raw-row pins (TestContractPin_MediaListBothBranches →
+rawRowKeys) whitelist the key. Nothing removed, nothing renamed.
+
+## 2026-10-30 — episode collections (additive widening, plan item 2/2.3/2.4)
+
+Two shapes widen, both ADDITIVE — nothing removed, nothing renamed:
+
+- **Collections envelope** (`GET /api/collections/:id` → envelopeKeys)
+  gains `rule_match_mode` (smart-collection match semantics, 'any'
+  legacy default | 'all'). Static collections emit the default value.
+- **BookInfo** (every card surface → bookInfoKeys) gains four STABLE
+  keys — `is_episode`, `feed_title`, `download_state`, `hidden` — the
+  episode-collections extras. Adapters that predate episode awareness
+  emit their zero values (`false`/`""`), matching the
+  viewer_percentage precedent (nil-representable, key always present).
+
+## 2026-10-30 — collections envelope gains filter_tree (additive widening, item 3.2)
+
+`GET /api/collections/:id` adds `filter_tree` to the envelope: the
+unified filter tree when the smart collection carries one (item 3.2 —
+tree-driven membership), `null` for legacy/static collections. The
+tree is the same qf DSL as saved filters (`{scope?, mode, children}`).
+Nothing removed, nothing renamed.
+
+## 2026-10-30 — rules endpoints RETIRED (filter-system unification, item 3.4)
+
+The per-rule CRUD endpoints and the preview endpoint are DELETED — the
+unified filter tree (item 3.2) is the only smart-membership authoring
+surface, and the rules PAGE is now the filter builder in the
+collection's scope:
+
+- `POST /api/collections/:id/rules`, `PUT/DELETE
+  /api/collections/:id/rules/:ruleId` — gone (legacy rule arrays still
+  EVALUATE until the collection is rewritten; conversions run at the
+  builder's first Apply via RulesToFilterTree).
+- `POST /api/collections/test-rules` — gone (the builder's live count
+  via /api/media-items/filter-count replaces the preview).
+- `POST /api/collections/preview` STAYS (the dashboard
+  custom-section-builder still consumes it — flagged for its own
+  migration, not silently broken).
+- Legacy compatibility KEPT: `auto_assign_rules` still round-trips in
+  the collections envelope; `PUT /api/collections/:id/match-mode`
+  still serves rules-array collections.
+
+Test-side deletions follow the asymmetry rule: subtests exercising the
+deleted endpoints are gone; their surviving SEMANTICS are pinned in
+their tree form (TestSmartCollections rewritten on trees,
+TestEpisodeCollectionsFilterTree, TestPodcastSmartCollections).
+
+## 2026-10-31 — collections RLS: cross-user detail/writes tighten to 404/no-op
+
+`collections` and `collection_items` joined the RLS-governed tables
+(policies `p_collections_*` / `p_collection_items_*`, predicates
+`app_can_read_collection` / `app_can_write_collection` — the same
+doctrine as the media_items family lens, 2026-10-29). Contract
+changes:
+
+- `GET /api/collections/:id` for another user's collection: was 200
+  (by-id readability), now 404 (row invisible).
+- `PUT /api/collections/:id` cross-user: was ungated success (no
+  app-level ownership check), now 404 — the RLS update touches no
+  rows, and the handler maps ErrNoRows to 404.
+- `POST /api/collections/:id/books` cross-user: was ungated, now 404
+  (rejectSmartMutation's GetCollectionData is RLS-invisible).
+- `DELETE /api/collections/:id` cross-user: still 204, but now a true
+  DB-level no-op (zero rows deleted) instead of an app-level free hit.
+- `PUT /api/collections/:id/view-settings` cross-user: 404 (the owner
+  gate stays as defense-in-depth; the DB is enforcement of record).
+- `PUT .../match-mode` and `.../items/order` keep their app-level 403
+  gates; the DB now backs them.
+- Reads: owner or admin (policy escape); non-owner non-admin users see
+  nothing, anonymous included. Future sharing rides the predicate
+  functions — one EXISTS branch in app_can_read/write_collection, no
+  policy migration.
+
+Test-side: TestCollectionsRLS pins the cross-user matrix end-to-end;
+fixture writes to non-system collections run as the row's OWNER
+(scopeExecAs/ownerSeedCtx) because the admin write-escape covers
+system collections only.
+
+## 2026-10-31 — legacy collection rules surface RETIRED (tree-only storage)
+
+The pre-tree rules machinery is gone — no production row ever carried a
+rules array or a non-default match mode (verified in the dev DB), the
+web never reads the envelope keys, and the Android app consumes none of
+it. Deletions:
+
+- `PUT /api/collections/:id/match-mode` — endpoint retired (it had no
+  doc page). The tree's root `mode` owns any|all semantics now.
+- Envelope: `auto_assign_rules` and `rule_match_mode` are REMOVED from
+  the create/list/detail/update payloads. This entry retracts the
+  2026-10-30 pins that listed them. `collection_type` and `is_smart`
+  stay.
+- Request: `auto_assign_rules`/`rule_match_mode` are no longer accepted
+  on create/update. A create carrying `auto_assign_rules` answers 400
+  naming the retirement — the custom-section builder's legacy save
+  (already failing to bind, verified live) fails loudly instead of
+  silently creating a hollow static section.
+- Storage: the `collections.auto_assign_rules` and
+  `collections.rule_match_mode` columns are DROPPED (idempotent
+  `DROP COLUMN IF EXISTS` in schema.sql).
+- Service: the rules CRUD layer (AddRule/SetRuleEnabled/RemoveRule +
+  validateRule + the UpdateCollectionRulesJSON query),
+  ParseRules/EnabledRules/IsSmartCollection, RulesToFilterTree (+ the
+  SSR preload branch), and the legacy branch of EvaluateSmartMembership
+  are deleted. IsCollectionSmart = tree presence.
+
+KEPT (live consumers): `POST /api/collections/preview` still accepts
+`rules` and evaluates them in memory (the custom-section builder's
+preview button); `include_auto` on the list stays (the Android app
+sends it). The builder's SAVE remains the flagged §12 tree migration.
+
+## 2026-10-31 — custom-section builder save: the flagged tree migration lands
+
+Supersedes the same-day retirement entry's loud-400 stance for the
+builder's payload. `POST /api/collections` now ACCEPTS the builder's
+shape and adapts it at write time:
+
+- `auto_assign_rules` (JSON-encoded array — string or bare) converts to
+  a filter tree via the restored `RulesToFilterTree`; converted trees
+  validate under `ParseFilterTree`, so an unhonorable rule is a 400 at
+  write naming the problem.
+- `match_type` seeds the tree's root mode (`any` default).
+- `manual_book_ids` attach to the created collection after create
+  (unparsable ids skipped — the AddBooks leniency).
+- `show_on_dashboard` persists (new create field).
+- `library_id` remains selection context only (not stored).
+
+Dashboard sections: smart user sections evaluate their tree LIVE —
+`GetUserCollectionItems` merges stored membership with the composed
+engine's matches (deduped, library-scoped, capped at the section
+limit). This exposed and fixed a real wiring bug: every production
+`NewDashboardService` call omitted the search service, so tree
+evaluation through the dashboard could never work (the test harness
+wired it correctly — the gap was production-only). The dashboard
+handler and both inline sites now receive it, and evaluation failures
+log instead of silently emptying a section.
+
+## 2026-11-01 — feed operator vocabulary collapsed (Option A): one name per semantic
+
+The feed field's menu offered `is`/`isn't` alongside `in`/`isn't in` —
+exact synonyms (the clause compiles both pairs to the same `ANY`). The
+MENUS now offer `in`/`isn't in` (+ has/missing) only. This is an
+offered-vocabulary narrowing, not a parse change:
+
+- Stored trees and saved filters carrying `feed is X` / `feed is_not X`
+  PARSE AND EVALUATE FOREVER (`legacyFieldOps` accepts the synonyms at
+  validation — the stored wire is never invalidated).
+- The fields payload (the client's operator menus) no longer carries
+  the duplicates.
+- Client id-cells are uniform: every operator on an id-valued field
+  accepts a set, so all id cells say "pick one or more" and accumulate
+  picks — no per-operator placeholder/pick distinction to explain.
+
+## 2026-11-01 — id labels resolve at render (SSR embed): refresh keeps names
+
+Pages that render a filter tree embed the tree's id→name map at render
+time (`data-id-labels` on the builder mount) — the collection editor
+(stored tree) and the bookshelf (URL `qf`). Before this, refresh showed
+raw uuids: only ids persist in trees; the client's id→label cache dies
+with the page, and the quiet suggest fetch reused the MEDIA-CONSTRAINED
+facet, so an id whose media doesn't match the tree's sibling rules got
+no row and no label (live repro: `feed in [A,B,C] AND title contains
+Chapter` labeled one feed of three — two feeds had no "Chapter"
+episodes).
+
+- `SearchService.FilterTreeLabels` reads the ENTITY tables (no media
+  join, no tree constraint): podcast_feeds and collections are
+  RLS-governed through the scoped handle (invisible ids resolve to
+  nothing); libraries mirror the facet's visibility join; added_by is a
+  plain username lookup. A zero-episode feed resolves its title —
+  pinned (TestRulesPageEmbedsIdLabels), as is the hidden-library
+  non-resolution.
+- The bookshelf drawer also HYDRATES at component init now (no
+  data-tree mount → read the form): the parent's x-init
+  (initBookshelf) runs BEFORE the nested builder component exists
+  (Alpine initializes parents first), so its builderData() call was
+  null and a URL-qf refresh left the drawer with an empty seeded rule
+  while the grid stayed filtered. Init-order contract pinned in
+  filter_builder.test.ts.
+- The quiet suggest fetch remains only as a fallback for trees the SSR
+  never saw (client-side apply flows); embedded labels make it a no-op
+  on refresh.
+
+## 2026-11-01 — unset podcast order defaults to reversed (audiobook-first)
+
+The per-feed listening order (the D5 `podcast_order:<feedID>` KV shown
+in the Order dropdown) resolved its UNSET state to "native" — feed
+document order, which runs newest-first for classic RSS, so an
+audiobook feed played its LAST chapter first. The unset default is now
+`reverse` at every resolution site of that one setting: the podcast
+page list + dropdown (`podcastOrderForFeed`), the player's feed chain
+(reader.go), and the download worker (`orderOverrideFor`).
+
+- Stored overrides are untouched: anyone who picked an order keeps it
+  (native remains a valid, savable choice).
+- No ordering semantics changed — `OrderEpisodes` and the four-option
+  vocabulary are as they were; only the unset fallback flipped.
+- Pinned: the player sequence config carries `"order":"reverse"` and
+  feed position 1 plays LAST for an unset feed; the podcast page's
+  dropdown renders `reverse` selected (TestPodcastPlayerOrderControl,
+  replacing the former native pin).
+
+## 2026-11-01 — new podcast smart collections are BORN reversed
+
+`POST /api/collections` with an EPISODES-scope `filter_tree` seeds
+`view_settings.chain_order = "reverse"` on the new row (merge semantics:
+an explicit client `chain_order` is never clobbered). The dropdown's
+default selection and the chain/page behavior come from the STORED
+value — no resolution-logic changes; books-scope smart collections and
+statics get nothing seeded; existing collections are untouched.
+
+Pinned (TestSmartCollectionBornReversed): episodes seed, explicit-keep,
+books-scope no-seed, static no-seed; TestEpisodeCollectionsFilterTree's
+born-reversed chain/page pins replace the former native expectations
+(which changelog here per the expectation-change rule). One misplaced
+pin from 81e54f89 was corrected while flipping: the API envelope's
+match list is NOT chain-ordered and never was — order pins live on the
+SSR page and the reader chain.
+
+## 2026-11-01 — label endpoint replaces the suggest-facet fallback
+
+`GET /api/filter/labels?field=X&ids=a,b,c` names id-kind filter values
+from their source tables (entity rows only — no media join, no tree
+constraint; visibility/RLS ride the request identity). Non-id fields
+are a 400; ids truncate at the same 100 cap the SSR embed uses.
+
+The client's label resolution (`resolveIdLabels`/`resolveLabelsAsync`)
+now uses ONLY this endpoint. The former fallback — the quiet suggest
+fetch through the MEDIA-CONSTRAINED facet — is deleted: it labeled only
+the ids whose media matched the tree's sibling rules (the live feed
+repro), which mattered exactly where the SSR embed can't reach:
+applying a SAVED filter from a cold bookshelf. The apply flow now
+resolves through the endpoint and rebuilds the chips bar when labels
+land. Pinned (TestFilterLabelsEndpoint): a zero-episode feed resolves
+its title, comma-joined ids resolve together, an unsubscribed feed id
+stays unnamed (RLS), non-id fields are 400.
+
+## 2026-11-01 — the rules-Apply path seeds the reversed default too
+
+PUT /api/collections/:id carrying an EPISODES-scope filter_tree seeds
+`chain_order = "reverse"` when neither the request nor the stored
+view_settings names one — the creation seed's twin: a static collection
+becoming smart lands with the same default a born-smart one gets.
+Explicit values (client-sent or stored — e.g. a deliberate native)
+are never clobbered. Pinned (TestSmartCollectionBornReversed):
+apply-seeds-when-unordered, respects stored native, respects client
+native.
+
+TestPodcastSmartCollections' acceptance pin updated with the same flip:
+the briefing acceptance now starts from the page's FIRST card (feed
+position 5 — the born-reversed default's book order) and chains into
+feed position 1. The acceptance meaning is unchanged: play from what
+you see, chained back-to-back within the live briefing set.
+
+## 2026-11-01 — audio resume rewind setting
+
+- `GET /api/media-items/:id/chapters` gained one ADDITIVE response key:
+  `resume_rewind_seconds` (int, the requesting user's setting; 0 for
+  anonymous). Existing keys unchanged. The web player and the Android
+  app both gate the value through the 0/5/10 allow-list, so servers
+  without the key (absent → 0) degrade to exact resume.
+- `GET /api/auth/profile` now always carries
+  `audio_resume_rewind_seconds` (int, never omitted — 0 is a real
+  value, exact resume). `PUT /api/auth/profile` accepts the key as the
+  keep_screen_awake-style STRING ("0"/"5"/"10") so the profile form's
+  select can send an explicit Off; JSON clients that omit it leave the
+  column untouched, and anything outside {0,5,10} is 400.
+  Pinned (TestAudioResumeRewind): default 0, PUT "5" rides the
+  chapters payload, non-allow-listed rejections change nothing,
+  explicit "0" resets everywhere, omitted key leaves it untouched.

docs/developer/api/collections/add_auto_assign_rule.md

@@ -1,89 +0,0 @@
----
-method: POST
-path: /api/collections/:id/rules
-auth: jwt
-area: collections
----
-<!-- source: internal/handlers/collections.go:877 CollectionHandler.AddRule -->
-
-# Add Auto-Assign Rule
-
-Add an automatic book assignment rule to a collection.
-
-**Endpoint**: `POST /api/collections/{id}/rules`
-**Auth**: Required
-**Content-Type**: `application/json`
-
-## Path Parameters
-
-| Parameter | Type          | Required | Description     |
-| --------- | ------------- | -------- | --------------- |
-| id        | string (UUID) | Yes      | Collection UUID |
-
-## Request Body
-
-| Field    | Type          | Required | Description                                                                                                       |
-| -------- | ------------- | -------- | ----------------------------------------------------------------------------------------------------------------- |
-| field    | string        | Yes      | Field to match on (genre, author, series, language, publisher, copyright_year, tags)                              |
-| operator | string        | Yes      | Comparison operator (equals, not_equals, contains, not_contains, starts_with, ends_with, greater_than, less_than) |
-| value    | string/number | Yes      | Value to compare against                                                                                          |
-| priority | integer       | No       | Rule priority (1 = highest, default: 1)                                                                           |
-| enabled  | boolean       | No       | Whether rule is active (default: true)                                                                            |
-
-### Supported Fields
-
-| Field          | Type   | Description                           |
-| -------------- | ------ | ------------------------------------- |
-| genre          | string | Book genre                            |
-| author         | string | Book author                           |
-| series         | string | Book series name                      |
-| language       | string | Book language                         |
-| publisher      | string | Publisher name                        |
-| copyright_year | number | Publication year (numeric comparison) |
-| tags           | string | Book tags                             |
-
-### Supported Operators
-
-| Operator     | Type   | Description                           |
-| ------------ | ------ | ------------------------------------- |
-| equals       | all    | Exact match                           |
-| not_equals   | all    | Not equal                             |
-| contains     | string | Contains substring (case-insensitive) |
-| not_contains | string | Does not contain                      |
-| starts_with  | string | Starts with (case-insensitive)        |
-| ends_with    | string | Ends with (case-insensitive)          |
-| greater_than | number | Greater than                          |
-| less_than    | number | Less than                             |
-
-### Example Request
-
-```json example-request
-{
-  "field": "genre",
-  "operator": "equals",
-  "value": "Science Fiction",
-  "priority": 1,
-  "enabled": true
-}
-```
-
-## Response (201 Created)
-
-```json example-response
-{
-  "collection_id": null,
-  "is_smart": null,
-  "rule": null
-}
-```
-
-## Error Responses
-
-| Code | Description                         |
-| ---- | ----------------------------------- |
-| 400  | Invalid request (validation failed) |
-| 401  | Authentication required             |
-| 404  | Collection not found                |
-| 500  | Internal server error               |
-
-## Try It Out

docs/developer/api/collections/create_collection.md

@@ -23,7 +23,6 @@ Creating a collection whose name already exists for the user returns **409**.
   "color": "#00FF00",
   "icon": "📖",
   "collection_type": "static",
-  "auto_assign_rules": [],
   "view_settings": { "kobo": { "view_mode": "grid" } }
 }
 ```
@@ -39,7 +38,6 @@ Creating a collection whose name already exists for the user returns **409**.
   "color": "#00FF00",
   "icon": "📖",
   "collection_type": "static",
-  "auto_assign_rules": null,
   "view_settings": { "kobo": { "view_mode": "grid" } },
   "book_count": 0,
   "created_at": "2026-10-24 13:57:01.81234 +0000 UTC"
@@ -57,20 +55,13 @@ RFC3339. Treat it as display text; do not parse it as an ISO timestamp.
 | description | string | no | |
 | color | string | no | hex color, e.g. `"#FF5733"` |
 | icon | string | no | emoji |
-| collection_type | string | no | `"static"` (default) or `"smart"`; also implied by non-empty `auto_assign_rules` |
+| collection_type | string | no | `"static"` (default) or `"smart"`; also implied by a `filter_tree` |
-| auto_assign_rules | array | no | rule objects, see below |
 | view_settings | object | no | per-device display preferences |
-
+| filter_tree | object | no | the unified filter tree (qf DSL); presence makes the collection smart |
-### Auto-assign rule object
+| auto_assign_rules | string \| array | no | custom-section-builder payload — converted to `filter_tree` at write (see custom-section-builder doc) |
-
+| match_type | string | no | builder payload: `any` (default) or `all` — seeds the converted tree's root mode |
-| Field | Type | Required | Notes |
+| manual_book_ids | string\[\] | no | builder payload: attached to the collection after create |
-| ----- | ---- | -------- | ----- |
+| show_on_dashboard | boolean | no | pins the collection as a dashboard section |
-| id | string | no | rule id |
-| field | string | yes | one of `genre`, `series`, `author`, `language`, `publisher`, `copyright_year`, `tags` |
-| operator | string | yes | one of `equals`, `not_equals`, `contains`, `not_contains`, `starts_with`, `ends_with`, `greater_than`, `less_than` |
-| value | string | yes | value to compare against |
-| priority | number | no | 1-10; higher values take precedence |
-| enabled | boolean | no | absent = enabled (rules written before this field existed keep matching) |
 
 ## Errors
 

docs/developer/api/collections/get_collection.md

@@ -27,7 +27,6 @@ GET /api/collections/550e8400-e29b-41d4-a716-446655440000
   "description": "Books I want to read soon",
   "color": "#00FF00",
   "icon": "📖",
-  "auto_assign_rules": null,
   "view_settings": { "kobo": { "view_mode": "grid" } },
   "books": [],
   "book_count": 0,
@@ -57,8 +56,8 @@ GET /api/collections/550e8400-e29b-41d4-a716-446655440000
 | description | string | |
 | color | string | |
 | icon | string | |
-| auto_assign_rules | array \| null | |
 | view_settings | object | |
+| filter_tree | object \| null | the unified filter tree (qf DSL); null = static |
 | books | array | book cards in this collection |
 | book_count | number | length of `books` |
 | is_smart | boolean | |

docs/developer/api/collections/list_collections.md

@@ -29,7 +29,6 @@ GET /api/collections?include_auto=true&sort_by=name
       "description": "Books I want to read soon",
       "color": "#00FF00",
       "icon": "📖",
-      "auto_assign_rules": null,
       "created_at": "2026-10-24 13:57:01.81234 +0000 UTC",
       "book_count": 12,
       "is_smart": false
@@ -65,7 +64,6 @@ GET /api/collections?include_auto=true&sort_by=name
 | description | string | |
 | color | string | |
 | icon | string | |
-| auto_assign_rules | array \| null | raw rules JSON; non-null implies smart |
 | created_at | string | Go `time.String()` format |
 | book_count | number | live count; respects `library_id` for smart collections |
 | is_smart | boolean | |

docs/developer/api/collections/remove_auto_assign_rule.md

@@ -1,46 +0,0 @@
----
-method: DELETE
-path: /api/collections/:id/rules/:ruleId
-auth: jwt
-area: collections
----
-<!-- source: internal/handlers/collections.go:960 CollectionHandler.RemoveRule -->
-
-# Remove Auto-Assign Rule
-
-Remove an automatic book assignment rule from a collection.
-
-**Endpoint**: `DELETE /api/collections/{collectionId}/rules/{ruleId}`
-**Auth**: Required
-**Content-Type**: `application/json`
-
-## Path Parameters
-
-| Parameter    | Type          | Required | Description     |
-| ------------ | ------------- | -------- | --------------- |
-| collectionId | string (UUID) | Yes      | Collection UUID |
-| ruleId       | string (UUID) | Yes      | Rule UUID       |
-
-## Response (204 No Content)
-
-Rule deleted successfully. No response body.
-
-```json example-response
-{
-  "collection_id": "550e8400-e29b-41d4-a716-446655440000",
-  "is_smart": true
-}
-```
-
-`is_smart` reflects the collection state AFTER removal — removing the
-last rule converts a smart collection back to static.
-
-## Error Responses
-
-| Code | Description                  |
-| ---- | ---------------------------- |
-| 401  | Authentication required      |
-| 404  | Collection or rule not found |
-| 500  | Internal server error        |
-
-## Try It Out

docs/developer/api/collections/set_chain_order.md

@@ -0,0 +1,48 @@
+---
+method: PUT
+path: /api/collections/:id/view-settings
+auth: jwt
+area: collections
+---
+<!-- source: internal/handlers/collections.go CollectionHandler.SetCollectionChainOrder -->
+
+# Set Collection Chain Order
+
+Sets the curated listening order for the collection's episode sequence
+(plan §10 R2): the chain producer (`buildSmartCollectionSequenceContext`)
+plays `Play as Sequence` in this order. The key merges into the
+collection's `view_settings` JSONB — playlist toggles that share the
+document survive the write. Owner-only.
+
+## Example request
+
+```json example-request
+{
+  "chain_order": "newest"
+}
+```
+
+## Example response
+
+```json example-response
+{
+  "id": "550e8400-e29b-41d4-a716-446655440000",
+  "chain_order": "newest",
+  "view_settings": { "chain_order": "newest" }
+}
+```
+
+## Fields
+
+| Field | Type | Required | Notes |
+| ----- | ---- | -------- | ----- |
+| chain_order | string | Yes | One of `native`, `newest`, `oldest`, `reverse`; `native`/empty removes the stored key |
+
+## Errors
+
+| Status | Body | When |
+| ------ | ---- | ---- |
+| 400 | `{"error": "invalid collection id"}` | malformed id |
+| 400 | `{"error": "chain_order must be one of: native, newest, oldest, reverse"}` | unknown order |
+| 404 | `{"error": "collection not found"}` | unknown id |
+| 401 / 403 | — | missing JWT / not the owner |

docs/developer/api/collections/test_rule.md

@@ -1,79 +0,0 @@
----
-method: POST
-path: /api/collections/test-rules
-auth: jwt
-area: collections
----
-<!-- source: internal/handlers/collections.go:996 CollectionHandler.TestRules -->
-
-# Test Rule
-
-Test which books would match given rules without saving.
-
-**Endpoint**: `POST /api/collections/test-rules`
-**Auth**: Required
-**Content-Type**: `application/json`
-
-## Request Body
-
-| Field | Type  | Required | Description                   |
-| ----- | ----- | -------- | ----------------------------- |
-| rules | array | Yes      | Array of rule objects to test |
-
-### Rule Object
-
-| Field    | Type          | Required | Description                                                                                                       |
-| -------- | ------------- | -------- | ----------------------------------------------------------------------------------------------------------------- |
-| field    | string        | Yes      | Field to match on (genre, author, series, language, publisher, copyright_year, tags)                              |
-| operator | string        | Yes      | Comparison operator (equals, not_equals, contains, not_contains, starts_with, ends_with, greater_than, less_than) |
-| value    | string/number | Yes      | Value to compare against                                                                                          |
-
-### Example Request
-
-```json example-request
-{
-  "rules": [
-    {
-      "field": "genre",
-      "operator": "equals",
-      "value": "Science Fiction"
-    },
-    {
-      "field": "author",
-      "operator": "contains",
-      "value": "Asimov"
-    }
-  ]
-}
-```
-
-## Response (200 OK)
-
-```json example-response
-{
-  "matches": [
-    {
-      "media_item_id": "660e8400-e29b-41d4-a716-446655440000",
-      "title": "Foundation",
-      "author": "Isaac Asimov",
-      "cover_image_path": "/covers/foundation.jpg",
-      "match_reason": "Matched rule: genre equals Science Fiction"
-    }
-  ],
-  "total": 42
-}
-```
-
-## Use Case
-
-Test rules before creating a collection to verify correct book matching. This endpoint shows which books would be added without actually modifying any collections.
-
-## Error Responses
-
-| Code | Description                         |
-| ---- | ----------------------------------- |
-| 400  | Invalid request (validation failed) |
-| 401  | Authentication required             |
-| 500  | Internal server error               |
-
-## Try It Out

docs/developer/api/collections/update_collection.md

@@ -24,7 +24,6 @@ set `view_settings.hide_finished` / `view_settings.show_up_next` directly.
   "description": "Updated description",
   "color": "#33CCFF",
   "icon": "📚",
-  "auto_assign_rules": [],
   "view_settings": { "hide_finished": true },
   "hide_finished": ["true"],
   "show_up_next": ["true"]
@@ -41,7 +40,6 @@ set `view_settings.hide_finished` / `view_settings.show_up_next` directly.
   "description": "Updated description",
   "color": "#33CCFF",
   "icon": "📚",
-  "auto_assign_rules": null,
   "view_settings": { "hide_finished": true, "show_up_next": true },
   "created_at": "2026-10-24 13:57:01.81234 +0000 UTC"
 }
@@ -57,8 +55,8 @@ set `view_settings.hide_finished` / `view_settings.show_up_next` directly.
 | description | string | no | |
 | color | string | no | |
 | icon | string | no | |
-| auto_assign_rules | array | no | replaces the whole rule set (rule object shape: see Create Collection) |
 | view_settings | object | no | merged into stored settings |
+| filter_tree | object | no | replaces the stored filter tree (omitting it keeps the stored one) |
 | hide_finished | array | no | single-element string array `["true"]`/`["false"]`; merged into view_settings |
 | show_up_next | array | no | same shape as `hide_finished` |
 

docs/developer/api/collections/update_rule.md

@@ -1,45 +0,0 @@
----
-method: PUT
-path: /api/collections/:id/rules/:ruleId
-auth: jwt
-area: collections
----
-<!-- source: internal/handlers/collections.go:922 CollectionHandler.UpdateRule (SetRuleEnabled) -->
-
-# Update Collection Rule
-
-Despite the route name, this endpoint **toggles a rule's enabled state only**.
-It does not update the rule's field/operator/value — rules are replaced wholesale
-via [Create Collection](create_collection.md) (`auto_assign_rules`) or added via
-the rules endpoint.
-
-## Example request
-
-```json example-request
-{
-  "enabled": false
-}
-```
-
-## Example response
-
-```json example-response
-{
-  "collection_id": "550e8400-e29b-41d4-a716-446655440000",
-  "is_smart": true
-}
-```
-
-## Request fields
-
-| Field | Type | Required | Notes |
-| ----- | ---- | -------- | ----- |
-| enabled | boolean | yes | toggles the rule on/off |
-
-## Errors
-
-| Status | Body | When |
-| ------ | ---- | ---- |
-| 400 | `{"error": "invalid collection id"}` / `{"error": "rule id required"}` / `{"error": "invalid request"}` | malformed |
-| 404 | `{"error": "rule not found"}` | no such rule |
-| 500 | `{"error": "..."}` | unexpected server error |

docs/developer/api/custom-section-builder.md

@@ -68,6 +68,15 @@ Creates a new custom collection with filter rules and/or manual book selection.
 
 **Request Body:**
 
+**Save contract (2026-10-31 tree migration):** the rules array travels
+as a JSON-encoded string and the server converts it to a filter tree at
+write time — `match_type` seeds the tree's root mode (`any` default),
+`manual_book_ids` attach to the created collection, and
+`show_on_dashboard` persists. `library_id` is selection context only
+(sections scope to the dashboard's active library at render). Invalid
+rules are a 400 at write naming the problem; the section renders stored
+picks + the tree's live matches, deduped.
+
 ```json
 {
   "name": "My Custom Section",

docs/developer/api/libraries/add_library_folder.md

@@ -41,7 +41,8 @@ Add a folder to an existing library (Admin only).
   "id": "uuid",
   "library_id": "uuid",
   "folder_path": "/path/to/comics",
-  "created_at": "2026-01-31T10:00:00Z"
+  "created_at": "2026-01-31T10:00:00Z",
+  "purpose": "scan"
 }
 ```
 

docs/developer/api/libraries/create_library.md

@@ -42,7 +42,8 @@ Create a new library (Admin only).
   "created_at": "2026-01-31T10:00:00Z",
   "created_by_admin_id": null,
   "library_type_id": null,
-  "updated_at": null
+  "updated_at": null,
+  "allow_member_feeds": false
 }
 ```
 

docs/developer/api/libraries/get_library.md

@@ -43,7 +43,8 @@ Authorization: Bearer eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9...
   "created_at": null,
   "created_by_admin_id": null,
   "library_type_id": null,
-  "updated_at": null
+  "updated_at": null,
+  "allow_member_feeds": false
 }
 ```
 

docs/developer/api/libraries/get_library_folders.md

@@ -18,7 +18,8 @@ Lists the filesystem folders configured for a library.
     "id": "aa0e8400-e29b-41d4-a716-446655440000",
     "library_id": "990e8400-e29b-41d4-a716-446655440000",
     "folder_path": "/books/manga",
-    "created_at": "2026-01-01T00:00:00Z"
+    "created_at": "2026-01-01T00:00:00Z",
+    "purpose": "scan"
   }
 ]
 ```

docs/developer/api/libraries/get_visible_libraries.md

@@ -42,7 +42,8 @@ A top-level JSON **array** of library rows:
     "updated_at": "2026-01-31T10:00:00Z",
     "type_name": "ebooks",
     "type_description": "Ebook libraries",
-    "is_visible": true
+    "is_visible": true,
+    "allow_member_feeds": false
   }
 ]
 ```

docs/developer/api/libraries/update_library.md

@@ -46,7 +46,8 @@ Update a library's information.
   "created_at": "2026-02-08T10:00:00Z",
   "updated_at": "2026-02-08T11:00:00Z",
   "created_by_admin_id": null,
-  "description": null
+  "description": null,
+  "allow_member_feeds": false
 }
 ```
 

docs/developer/api/media-items/create_media_item.md

@@ -164,7 +164,23 @@ Tags and contributors are automatically normalized:
   "summary": null,
   "total_characters": null,
   "volume": null,
-  "web_url": null
+  "web_url": null,
+  "source": null,
+  "source_feed_id": null,
+  "feed_position": null,
+  "episode_number": null,
+  "season_number": null,
+  "episode_guid": null,
+  "enclosure_url": null,
+  "chapters_url": null,
+  "transcript_url": null,
+  "hosts": null,
+  "guests": null,
+  "declared_duration_seconds": null,
+  "download_state": null,
+  "download_attempts": null,
+  "download_queue_pos": null,
+  "download_error": null
 }
 ```
 

docs/developer/api/media-items/search_media_items.md

@@ -231,7 +231,23 @@ Authorization: Bearer eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9...
     "total_characters": null,
     "updated_at": null,
     "volume": null,
-    "web_url": null
+    "web_url": null,
+    "source": null,
+    "source_feed_id": null,
+    "feed_position": null,
+    "episode_number": null,
+    "season_number": null,
+    "episode_guid": null,
+    "enclosure_url": null,
+    "chapters_url": null,
+    "transcript_url": null,
+    "hosts": null,
+    "guests": null,
+    "declared_duration_seconds": null,
+    "download_state": null,
+    "download_attempts": null,
+    "download_queue_pos": null,
+    "download_error": null
   }
 ]
 ```

docs/developer/api/media-items/update_media_item.md

@@ -128,7 +128,23 @@ Update media item metadata (Admin only).
   "tags_search": null,
   "total_characters": null,
   "volume": null,
-  "web_url": null
+  "web_url": null,
+  "source": null,
+  "source_feed_id": null,
+  "feed_position": null,
+  "episode_number": null,
+  "season_number": null,
+  "episode_guid": null,
+  "enclosure_url": null,
+  "chapters_url": null,
+  "transcript_url": null,
+  "hosts": null,
+  "guests": null,
+  "declared_duration_seconds": null,
+  "download_state": null,
+  "download_attempts": null,
+  "download_queue_pos": null,
+  "download_error": null
 }
 ```
 

docs/developer/api/users/get_profile.md

@@ -36,6 +36,8 @@ Authorization: Bearer eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9...
   "first_name": "John",
   "last_name": "Doe",
   "role": "user",
+  "default_audio_speed": 1.5,
+  "audio_resume_rewind_seconds": 5,
   "keep_screen_awake": true
 }
 ```

docs/developer/openapi.yaml

@@ -229,7 +229,6 @@ paths:
                 "color": "#00FF00",
                 "icon": "📖",
                 "collection_type": "static",
-                "auto_assign_rules": [],
                 "view_settings": { "kobo": { "view_mode": "grid" } }
               }
       responses:
@@ -277,33 +276,6 @@ paths:
       responses:
         '200':
           description: Success
-  "/api/collections/test-rules":
-    post:
-      tags: [collections]
-      summary: "Test Rule"
-      security:
-        - bearerAuth: []
-      requestBody:
-        content:
-          application/json:
-            example: |-
-              {
-                "rules": [
-                  {
-                    "field": "genre",
-                    "operator": "equals",
-                    "value": "Science Fiction"
-                  },
-                  {
-                    "field": "author",
-                    "operator": "contains",
-                    "value": "Asimov"
-                  }
-                ]
-              }
-      responses:
-        '200':
-          description: OK
   "/api/collections/{id}":
     delete:
       tags: [collections]
@@ -335,7 +307,6 @@ paths:
                 "description": "Updated description",
                 "color": "#33CCFF",
                 "icon": "📚",
-                "auto_assign_rules": [],
                 "view_settings": { "hide_finished": true },
                 "hide_finished": ["true"],
                 "show_up_next": ["true"]
@@ -409,38 +380,10 @@ paths:
       responses:
         '200':
           description: Success
-  "/api/collections/{id}/rules":
+  "/api/collections/{id}/view-settings":
-    post:
-      tags: [collections]
-      summary: "Add Auto-Assign Rule"
-      security:
-        - bearerAuth: []
-      requestBody:
-        content:
-          application/json:
-            example: |-
-              {
-                "field": "genre",
-                "operator": "equals",
-                "value": "Science Fiction",
-                "priority": 1,
-                "enabled": true
-              }
-      responses:
-        '201':
-          description: Created
-  "/api/collections/{id}/rules/{ruleId}":
-    delete:
-      tags: [collections]
-      summary: "Remove Auto-Assign Rule"
-      security:
-        - bearerAuth: []
-      responses:
-        '204':
-          description: No Content
     put:
       tags: [collections]
-      summary: "Update Collection Rule"
+      summary: "Set Collection Chain Order"
       security:
         - bearerAuth: []
       requestBody:
@@ -448,11 +391,11 @@ paths:
           application/json:
             example: |-
               {
-                "enabled": false
+                "chain_order": "newest"
               }
       responses:
         '200':
-          description: Success
+          description: OK
   "/api/conflicts":
     get:
       tags: [conflicts]

docs/index.md

@@ -17,6 +17,11 @@ Complete guide to Bookhoard documentation. Find what you need quickly.
 - **Sync Configuration**
   - [Universal Sync Guide](user/sync-guide.md) - Understanding sync, book matching, conflicts
 
+- **Filters**
+  - [Filters — the complete guide](user/filters/index.md) - rules, groups (ALL/ANY), saved filters, smart collections
+  - [Filter cookbook](user/filters/cookbook.md) - ready-made recipes, exact clicks
+  - [Every filter field](user/filters/fields.md) - the generated, always-current reference
+
 - **Frontend & Settings** (Coming Soon)
   - Frontend Guide
   - User Areas Guide

docs/user/filters/cookbook.md

@@ -0,0 +1,78 @@
+# Filter cookbook
+
+Real questions, exact clicks. Every recipe uses the same builder described in
+[the guide](index.md) — same fields, same rules, wherever filters live.
+
+Recipes marked **collection** become smart collections that stay current forever; the rest are
+drawer filters you can save with a name.
+
+---
+
+## "New chapters of 7th Son, automatically" — collection
+
+1. **Collections** → open (or create) a collection → **Rules**.
+2. The scope is already *Episodes*. Add a rule: **Feed** → **in** → pick
+   *7th Son: Book One* (it becomes a chip — hit the × if you pick wrong).
+3. Add a second rule: **Title** → **contains** → `Chapter`.
+4. The header should read **MATCH ALL** (it does by default — every rule must match).
+5. **Apply to Collection**. Done: new chapters appear on their own, and the listening order
+   starts at **Feed order (reversed)** — chapter 1 first.
+
+---
+
+## "Lewis Carroll or George Orwell" — drawer
+
+1. **Filters** on the All Books page. Add: **Author** → **contains** → `Carroll`.
+2. Add a second rule: **Author** → **contains** → `Orwell`.
+3. Click the group header **MATCH ALL** once — it flips to **MATCH ANY** (or click the `AND`
+   chip between the two rules, same effect).
+4. **Apply Filters**. Either author is enough.
+
+*Flip it back to ALL and you'd get only books they both wrote — which is none, and the counter
+will tell you so before you apply.*
+
+---
+
+## "Long fantasy books I haven't started" — drawer
+
+1. **Genre** → **is** → `fantasy`.
+2. **Page count** → **at least** → `400`.
+3. **My reading** → **Reading status** → **is** → *Unfinished* (or your app's wording — the
+   status list is the fixed dropdown on that field).
+4. **MATCH ALL** stays on: all three must hold.
+5. Apply — then **Save** it as *"Big unread fantasy"* and it's in the Load menu forever.
+
+---
+
+## "Downloaded podcast episodes I haven't played" — collection
+
+1. A collection with an Episodes scope (see the first recipe — drop the Title rule if you want
+   whole feeds).
+2. **Downloaded** → **is true**.
+3. **Played** → **is false**.
+4. **MATCH ALL** on. **Apply to Collection**.
+
+---
+
+## "Added to my library this month" — drawer
+
+1. **Date added** → **within the last** → `30` (the days box — the field's page in
+   [Every filter field](fields.md) has its exact questions).
+2. Optionally **MATCH ALL** with **Library** → **in** → your audiobooks library.
+3. Save as *"New arrivals"*.
+
+---
+
+## "A dashboard shelf that builds itself" — dashboard
+
+1. Create a collection with the rules above (or open one).
+2. On the dashboard, add a section pointed at that collection.
+3. The shelf re-fills itself as your library changes — no maintenance.
+
+---
+
+## Sharing a filtered view
+
+Apply any filter, then copy the address bar. The filter rides in the URL — send it to another
+user of your server and they open the same narrowed view (they see it through their own
+libraries, not yours).

docs/user/filters/fields.md

@@ -0,0 +1,755 @@
+<!-- GENERATED by TestFieldReferenceDocIsCurrent (internal/services/filter_docs_test.go)
+     from the filter registry — the same source the real menus render. DO NOT EDIT by hand:
+     regenerate with FILTER_DOCS_REGEN=1 go test ./internal/services -run TestFieldReferenceDocIsCurrent -->
+
+# Every filter field
+
+This is the complete list of everything the filters can ask about, generated straight from the
+same source the app's menus render — it cannot drift from what you see in the UI.
+
+Filters come in two worlds:
+
+- **Books** — rules about books and your reading of them.
+- **Episodes** — rules about podcast episodes (feeds, downloaded, played).
+
+Every rule is **a field + a question + your values**. The "Ask" column is the question;
+the "Your values" column is what you provide. The **default** question is what a new rule
+starts with.
+
+## Books
+
+### Book facts
+
+**Title** — Type words.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is | matches exactly |
+| isn't | matches anything but this |
+| is empty | — |
+
+**Subtitle** — Type words.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is | matches exactly |
+| isn't | matches anything but this |
+| is empty | — |
+
+**Description** — Type words.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is | matches exactly |
+| isn't | matches anything but this |
+| is empty | — |
+
+**Summary** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| has | has a value *(default)* |
+| missing | is empty or missing |
+
+**Language** — Type words.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is | matches exactly |
+| isn't | matches anything but this |
+| is empty | — |
+
+**Page count** — Type a number.
+
+| Ask | Means |
+| --- | --- |
+| between | falls between two numbers |
+| at least | is at least this *(default)* |
+| at most | is at most this |
+| is empty | — |
+
+**Chapter count** — Type a number.
+
+| Ask | Means |
+| --- | --- |
+| between | falls between two numbers |
+| at least | is at least this *(default)* |
+| at most | is at most this |
+| is empty | — |
+
+**Total characters** — Type a number.
+
+| Ask | Means |
+| --- | --- |
+| between | falls between two numbers |
+| at least | is at least this *(default)* |
+| at most | is at most this |
+| is empty | — |
+
+**Community rating** — Type a number.
+
+| Ask | Means |
+| --- | --- |
+| between | falls between two numbers |
+| at least | is at least this *(default)* |
+| at most | is at most this |
+| is empty | — |
+
+**Date added** — Type a number of days.
+
+| Ask | Means |
+| --- | --- |
+| within the last | — *(default)* |
+| between | falls between two numbers |
+| is empty | — |
+
+**Date imported** — Type a number of days.
+
+| Ask | Means |
+| --- | --- |
+| within the last | — *(default)* |
+| between | falls between two numbers |
+| is empty | — |
+
+**Web URL** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| has | has a value *(default)* |
+| missing | is empty or missing |
+
+**Metadata notes** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| has | has a value *(default)* |
+| missing | is empty or missing |
+
+**ISBN** — Pick from a dropdown of real names — picks become removable chips, and you can pick several.
+
+| Ask | Means |
+| --- | --- |
+| is | is one of the ones you pick |
+| isn't | matches anything but this |
+| has | has a value *(default)* |
+| missing | is empty or missing |
+
+**ASIN** — Pick from a dropdown of real names — picks become removable chips, and you can pick several.
+
+| Ask | Means |
+| --- | --- |
+| is | is one of the ones you pick |
+| isn't | matches anything but this |
+| has | has a value *(default)* |
+| missing | is empty or missing |
+
+**File size** — Type a number.
+
+| Ask | Means |
+| --- | --- |
+| between | falls between two numbers |
+| at least | is at least this *(default)* |
+| at most | is at most this |
+| is empty | — |
+
+**File availability** — Choose from a fixed list of options.
+
+| Ask | Means |
+| --- | --- |
+| is | matches exactly *(default)* |
+| isn't | matches anything but this |
+
+**Explicit** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| is true | is switched on *(default)* |
+| is false | is switched off |
+
+**Abridged** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| is true | is switched on *(default)* |
+| is false | is switched off |
+
+**Manually edited** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| is true | is switched on *(default)* |
+| is false | is switched off |
+
+**Cover** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| has | has a value *(default)* |
+| missing | is empty or missing |
+
+**Added by** — Pick from a dropdown of real names — picks become removable chips, and you can pick several.
+
+| Ask | Means |
+| --- | --- |
+| is | is one of the ones you pick |
+| has | has a value *(default)* |
+| missing | is empty or missing |
+
+### Comics & manga
+
+**Series** — Type words.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is | matches exactly |
+| isn't | matches anything but this |
+| is empty | — |
+
+**Series number** — Type a number.
+
+| Ask | Means |
+| --- | --- |
+| between | falls between two numbers |
+| at least | is at least this *(default)* |
+| at most | is at most this |
+| is empty | — |
+
+**Series count** — Type a number.
+
+| Ask | Means |
+| --- | --- |
+| between | falls between two numbers |
+| at least | is at least this *(default)* |
+| at most | is at most this |
+| is empty | — |
+
+**Volume** — Type a number.
+
+| Ask | Means |
+| --- | --- |
+| between | falls between two numbers |
+| at least | is at least this *(default)* |
+| at most | is at most this |
+| is empty | — |
+
+**Format group** — Pick from a dropdown of real names — picks become removable chips, and you can pick several.
+
+| Ask | Means |
+| --- | --- |
+| is | is one of the ones you pick *(default)* |
+| isn't | matches anything but this |
+
+**File type (mime)** — Pick from a dropdown of real names — picks become removable chips, and you can pick several.
+
+| Ask | Means |
+| --- | --- |
+| is | is one of the ones you pick *(default)* |
+| isn't | matches anything but this |
+
+**Reflowable** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| is true | is switched on *(default)* |
+| is false | is switched off |
+
+**Fixed layout** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| is true | is switched on *(default)* |
+| is false | is switched off |
+
+**Black & white** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| is true | is switched on *(default)* |
+| is false | is switched off |
+
+**Manga type** — Pick from a dropdown of real names — picks become removable chips, and you can pick several.
+
+| Ask | Means |
+| --- | --- |
+| is | is one of the ones you pick *(default)* |
+| isn't | matches anything but this |
+| is empty | — |
+
+**Reading direction** — Pick from a dropdown of real names — picks become removable chips, and you can pick several.
+
+| Ask | Means |
+| --- | --- |
+| is | is one of the ones you pick *(default)* |
+| isn't | matches anything but this |
+| is empty | — |
+
+### Library
+
+**Library** — Pick from a dropdown of real names — picks become removable chips, and you can pick several.
+
+| Ask | Means |
+| --- | --- |
+| is | is one of the ones you pick *(default)* |
+| isn't | matches anything but this |
+
+### My reading
+
+**Reading status** — Choose from a fixed list of options.
+
+| Ask | Means |
+| --- | --- |
+| is | matches exactly *(default)* |
+
+**Progress percent** — Type a number.
+
+| Ask | Means |
+| --- | --- |
+| between | falls between two numbers |
+| at least | is at least this *(default)* |
+| at most | is at most this |
+| is empty | — |
+
+**Last read** — Type a number of days.
+
+| Ask | Means |
+| --- | --- |
+| within the last | — *(default)* |
+| between | falls between two numbers |
+| is empty | — |
+
+**My rating** — Type a number.
+
+| Ask | Means |
+| --- | --- |
+| is | matches exactly |
+| at least | is at least this *(default)* |
+| at most | is at most this |
+| unrated | — |
+
+**Has review** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| has | has a value *(default)* |
+| missing | is empty or missing |
+
+**Notes** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| has | has a value *(default)* |
+| missing | is empty or missing |
+| within the last | — |
+
+**Highlights** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| has | has a value *(default)* |
+| missing | is empty or missing |
+| within the last | — |
+
+**Note text** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is empty | — |
+
+**Highlight text** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is empty | — |
+
+**Time spent (minutes)** — Type a number.
+
+| Ask | Means |
+| --- | --- |
+| between | falls between two numbers |
+| at least | is at least this *(default)* |
+| at most | is at most this |
+
+**Pages read** — Type a number.
+
+| Ask | Means |
+| --- | --- |
+| between | falls between two numbers |
+| at least | is at least this *(default)* |
+| at most | is at most this |
+
+**In collection** — Pick from a dropdown of real names — picks become removable chips, and you can pick several.
+
+| Ask | Means |
+| --- | --- |
+| in | is one of the ones you pick (any one is enough) *(default)* |
+| isn't in | is none of the ones you pick |
+| in any | is in at least one of the collections you pick |
+| in none | — |
+
+### People
+
+**Author** — Type words.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is | matches exactly |
+| isn't | matches anything but this |
+| is empty | — |
+
+**Narrator** — Type words.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is | matches exactly |
+| isn't | matches anything but this |
+| is empty | — |
+
+**Contributor** — Type words.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is | matches exactly |
+| isn't | matches anything but this |
+| is empty | — |
+
+### Publishing
+
+**Publisher** — Type words.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is | matches exactly |
+| isn't | matches anything but this |
+| is empty | — |
+
+**Imprint** — Type words.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is | matches exactly |
+| isn't | matches anything but this |
+| is empty | — |
+
+**Edition** — Type words.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is | matches exactly |
+| isn't | matches anything but this |
+| is empty | — |
+
+**Published year** — Type a number.
+
+| Ask | Means |
+| --- | --- |
+| between | falls between two numbers |
+| at least | is at least this *(default)* |
+| at most | is at most this |
+| is empty | — |
+
+**Published date** — Type a number of days.
+
+| Ask | Means |
+| --- | --- |
+| within the last | — *(default)* |
+| between | falls between two numbers |
+| is empty | — |
+
+### Topics
+
+**Genre** — Type words.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is | matches exactly |
+| isn't | matches anything but this |
+| is empty | — |
+
+**Tags** — Type words.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is | matches exactly |
+| isn't | matches anything but this |
+| is empty | — |
+
+**Story arc** — Type words.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is | matches exactly |
+| isn't | matches anything but this |
+| is empty | — |
+
+**Age rating** — Pick from a dropdown of real names — picks become removable chips, and you can pick several.
+
+| Ask | Means |
+| --- | --- |
+| is | is one of the ones you pick *(default)* |
+| isn't | matches anything but this |
+| is empty | — |
+
+## Podcast episodes
+
+### Book facts
+
+**Title** — Type words.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is | matches exactly |
+| isn't | matches anything but this |
+| is empty | — |
+
+**Description** — Type words.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is | matches exactly |
+| isn't | matches anything but this |
+| is empty | — |
+
+**Language** — Type words.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is | matches exactly |
+| isn't | matches anything but this |
+| is empty | — |
+
+**Chapter count** — Type a number.
+
+| Ask | Means |
+| --- | --- |
+| between | falls between two numbers |
+| at least | is at least this *(default)* |
+| at most | is at most this |
+| is empty | — |
+
+**Date added** — Type a number of days.
+
+| Ask | Means |
+| --- | --- |
+| within the last | — *(default)* |
+| between | falls between two numbers |
+| is empty | — |
+
+**Date imported** — Type a number of days.
+
+| Ask | Means |
+| --- | --- |
+| within the last | — *(default)* |
+| between | falls between two numbers |
+| is empty | — |
+
+**File size** — Type a number.
+
+| Ask | Means |
+| --- | --- |
+| between | falls between two numbers |
+| at least | is at least this *(default)* |
+| at most | is at most this |
+| is empty | — |
+
+**File availability** — Choose from a fixed list of options.
+
+| Ask | Means |
+| --- | --- |
+| is | matches exactly *(default)* |
+| isn't | matches anything but this |
+
+**Explicit** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| is true | is switched on *(default)* |
+| is false | is switched off |
+
+### Comics & manga
+
+**File type (mime)** — Pick from a dropdown of real names — picks become removable chips, and you can pick several.
+
+| Ask | Means |
+| --- | --- |
+| is | is one of the ones you pick *(default)* |
+| isn't | matches anything but this |
+
+### Episodes
+
+**Feed** — Pick from a dropdown of real names — picks become removable chips, and you can pick several.
+
+| Ask | Means |
+| --- | --- |
+| in | is one of the ones you pick (any one is enough) *(default)* |
+| isn't in | is none of the ones you pick |
+| has | has a value |
+| missing | is empty or missing |
+
+**Downloaded** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| is true | is switched on *(default)* |
+| is false | is switched off |
+
+**Played** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| is true | is switched on |
+| is false | is switched off *(default)* |
+
+### Library
+
+**Library** — Pick from a dropdown of real names — picks become removable chips, and you can pick several.
+
+| Ask | Means |
+| --- | --- |
+| is | is one of the ones you pick *(default)* |
+| isn't | matches anything but this |
+
+### My reading
+
+**Progress percent** — Type a number.
+
+| Ask | Means |
+| --- | --- |
+| between | falls between two numbers |
+| at least | is at least this *(default)* |
+| at most | is at most this |
+| is empty | — |
+
+**Last read** — Type a number of days.
+
+| Ask | Means |
+| --- | --- |
+| within the last | — *(default)* |
+| between | falls between two numbers |
+| is empty | — |
+
+**My rating** — Type a number.
+
+| Ask | Means |
+| --- | --- |
+| is | matches exactly |
+| at least | is at least this *(default)* |
+| at most | is at most this |
+| unrated | — |
+
+**Has review** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| has | has a value *(default)* |
+| missing | is empty or missing |
+
+**Notes** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| has | has a value *(default)* |
+| missing | is empty or missing |
+| within the last | — |
+
+**Highlights** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| has | has a value *(default)* |
+| missing | is empty or missing |
+| within the last | — |
+
+**Note text** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is empty | — |
+
+**Highlight text** — No value needed — the question is the whole rule.
+
+| Ask | Means |
+| --- | --- |
+| contains | contains these words *(default)* |
+| doesn't contain | does NOT contain these words |
+| is empty | — |
+
+**Time spent (minutes)** — Type a number.
+
+| Ask | Means |
+| --- | --- |
+| between | falls between two numbers |
+| at least | is at least this *(default)* |
+| at most | is at most this |
+
+**In collection** — Pick from a dropdown of real names — picks become removable chips, and you can pick several.
+
+| Ask | Means |
+| --- | --- |
+| in | is one of the ones you pick (any one is enough) *(default)* |
+| isn't in | is none of the ones you pick |
+| in any | is in at least one of the collections you pick |
+| in none | — |
+
+### Publishing
+
+**Published year** — Type a number.
+
+| Ask | Means |
+| --- | --- |
+| between | falls between two numbers |
+| at least | is at least this *(default)* |
+| at most | is at most this |
+| is empty | — |
+
+**Published date** — Type a number of days.
+
+| Ask | Means |
+| --- | --- |
+| within the last | — *(default)* |
+| between | falls between two numbers |
+| is empty | — |
+
+---
+
+*Anything here that doesn't match what the app actually shows is a bug — this page is generated
+from the app's own menus and a test fails if they ever disagree.*

docs/user/filters/index.md

@@ -0,0 +1,91 @@
+# Filters — the complete guide
+
+Filters answer one question: *of everything in my library, show me the ones that match.*
+
+You'll meet the same filter system in three places:
+
+| Place | What it's for |
+| --- | --- |
+| **The Filters drawer** (All Books / Podcasts pages) | A quick, temporary narrowing of what you see. Save it if you want to keep it. |
+| **Smart collections** (a collection's **Rules** page) | A filter that *lives* — the collection's contents update automatically forever. |
+| **Dashboard sections** | The same rules, feeding a shelf on your home screen. |
+
+One system, one set of skills, three payoffs.
+
+## Two worlds: Books and Episodes
+
+Rules about **books** (authors, series, your reading progress) and rules about **podcast episodes**
+(feeds, downloaded, played) are two different worlds. The field menus only show the fields that
+make sense where you are — you'll never see a podcast field on the All Books page. The complete
+list of every field is in [Every filter field](fields.md).
+
+## Anatomy of a rule
+
+Every rule is **a field + a question + your values**:
+
+> **Author** *contains* `asc`
+
+> **Feed** *in* `7th Son: Book One`
+
+> **Page count** *at least* `400`
+
+The field is *what* you're asking about. The question (contains, at least, is empty…) is *how*.
+Your values are *what counts as a match*. The complete questions per field are in
+[Every filter field](fields.md).
+
+## Picking values
+
+Two kinds of value boxes:
+
+- **Dropdown boxes** (Feed, Library, Collection, Added by): start typing and real names appear,
+  with live counts of how much matches. Click to pick — each pick becomes a **chip** with a little
+  **×** to remove it. Pick as many as you like; *any one of them* matching is enough.
+- **Typed boxes** (Author, Title, Page count…): just type. Several words means "contains these
+  words"; number boxes take numbers; date boxes take a number of days.
+
+The dropdown's counts are **live under your other rules**: if you've already narrowed to
+*Title contains Chapter*, the feed menu shows how many Chapter-titled episodes each of your feeds
+has. Rules already picked disappear from their own menu (removing a chip brings them back).
+
+## Groups: ALL and ANY
+
+When you have several rules, the group header — **MATCH ALL** or **MATCH ANY** — decides how they
+combine:
+
+- **MATCH ALL** — *every* rule must match. (The default, and what you want almost always.)
+- **MATCH ANY** — *at least one* rule matching is enough. This is how you ask for
+  "Lewis Carroll **or** George Orwell".
+
+**Click the header to switch it.** It says what it does, and clicking flips it.
+
+`+ group` adds a nested group with its own header (**GROUP ALL / GROUP ANY**), so bigger shapes
+build naturally:
+
+> Match ALL: Author *contains* `Carroll` **AND** GROUP ANY (Title contains `Alice` **OR** Title
+> contains `Wonderland`)
+
+For fine-tuning inside an ALL group, the small **AND / OR** chip between two rules mixes them —
+click `AND` and it becomes `OR`, joining those neighbors together. (The header switch and these
+chips are two different tools: the header sets the group's overall combine; a chip makes just two
+neighboring rules special.)
+
+## The live count
+
+While you build, the counter at the bottom — *N books match* — runs your rules against the real
+library. It's the exact number you'll get when you apply.
+
+## Defaults worth knowing
+
+- **Podcast listening order**: podcast pages and new podcast collections start at
+  **Feed order (reversed)** — book order, chapter 1 first. Change it per feed or collection from
+  the order dropdown; your choice sticks.
+- **Hidden libraries** (family lens) never match — a filter can't accidentally surface a library
+  that's hidden from you.
+- **Episodes that aren't downloaded** stay out of playback chains (they still appear in lists).
+
+## Keeping a filter
+
+- **Save** (drawer): name it; it appears in the **Load** menu from then on.
+- **Apply to Collection** (rules page): the filter becomes the smart collection — live forever.
+- Filtered views are **shareable**: the URL carries the filter, so a bookmark or sent link opens
+  the same view.

docs/user/searching-and-filtering.md

@@ -35,3 +35,12 @@ You can combine tags with other filters:
 - Tags use fuzzy matching - "Sci Fi" will match "Science Fiction"
 - Best matching tags appear first in results
 - Autocomplete shows existing tags as you type with relevance scores
+
+
+## The complete filter guide
+
+The filters described above are one system that also powers smart collections and dashboard
+sections — groups with ALL/ANY, saved filters, and every field listed. The full manual lives in
+[this folder](filters/index.md): the [guide](filters/index.md), the
+[cookbook](filters/cookbook.md) of ready-made recipes, and the generated
+[field reference](filters/fields.md).

internal/database/db.go

@@ -1,6 +1,6 @@
 // Code generated by sqlc. DO NOT EDIT.
 // versions:
-//   sqlc v1.31.1
+//   sqlc v1.31.0
 
 package database
 

internal/database/models.go

@@ -1,6 +1,6 @@
 // Code generated by sqlc. DO NOT EDIT.
 // versions:
-//   sqlc v1.31.1
+//   sqlc v1.31.0
 
 package database
 
@@ -34,7 +34,7 @@ type Collections struct {
 	Description        pgtype.Text        `db:"description" json:"description"`
 	Color              pgtype.Text        `db:"color" json:"color"`
 	Icon               pgtype.Text        `db:"icon" json:"icon"`
-	AutoAssignRules    []byte             `db:"auto_assign_rules" json:"auto_assign_rules"`
+	FilterTree         []byte             `db:"filter_tree" json:"filter_tree"`
 	ViewSettings       []byte             `db:"view_settings" json:"view_settings"`
 	ShowOnDashboard    pgtype.Bool        `db:"show_on_dashboard" json:"show_on_dashboard"`
 	QueryType          pgtype.Text        `db:"query_type" json:"query_type"`
@@ -150,6 +150,7 @@ type Libraries struct {
 	CreatedByAdminID pgtype.UUID        `db:"created_by_admin_id" json:"created_by_admin_id"`
 	CreatedAt        pgtype.Timestamptz `db:"created_at" json:"created_at"`
 	UpdatedAt        pgtype.Timestamptz `db:"updated_at" json:"updated_at"`
+	AllowMemberFeeds bool               `db:"allow_member_feeds" json:"allow_member_feeds"`
 }
 
 type LibraryFolders struct {
@@ -157,6 +158,7 @@ type LibraryFolders struct {
 	LibraryID  pgtype.UUID        `db:"library_id" json:"library_id"`
 	FolderPath string             `db:"folder_path" json:"folder_path"`
 	CreatedAt  pgtype.Timestamptz `db:"created_at" json:"created_at"`
+	Purpose    string             `db:"purpose" json:"purpose"`
 }
 
 type LibraryTypes struct {
@@ -338,6 +340,22 @@ type MediaItems struct {
 	OpfIdentifier           pgtype.Text        `db:"opf_identifier" json:"opf_identifier"`
 	OpfUuid                 pgtype.Text        `db:"opf_uuid" json:"opf_uuid"`
 	HashConfidence          pgtype.Text        `db:"hash_confidence" json:"hash_confidence"`
+	Source                  string             `db:"source" json:"source"`
+	SourceFeedID            pgtype.UUID        `db:"source_feed_id" json:"source_feed_id"`
+	EpisodeGuid             pgtype.Text        `db:"episode_guid" json:"episode_guid"`
+	FeedPosition            pgtype.Int4        `db:"feed_position" json:"feed_position"`
+	SeasonNumber            pgtype.Int4        `db:"season_number" json:"season_number"`
+	EpisodeNumber           pgtype.Int4        `db:"episode_number" json:"episode_number"`
+	EnclosureUrl            pgtype.Text        `db:"enclosure_url" json:"enclosure_url"`
+	DeclaredDurationSeconds pgtype.Int4        `db:"declared_duration_seconds" json:"declared_duration_seconds"`
+	TranscriptUrl           pgtype.Text        `db:"transcript_url" json:"transcript_url"`
+	ChaptersUrl             pgtype.Text        `db:"chapters_url" json:"chapters_url"`
+	Hosts                   []string           `db:"hosts" json:"hosts"`
+	Guests                  []string           `db:"guests" json:"guests"`
+	DownloadState           string             `db:"download_state" json:"download_state"`
+	DownloadError           pgtype.Text        `db:"download_error" json:"download_error"`
+	DownloadAttempts        int32              `db:"download_attempts" json:"download_attempts"`
+	DownloadQueuePos        pgtype.Int8        `db:"download_queue_pos" json:"download_queue_pos"`
 }
 
 type MediaNotes struct {
@@ -405,6 +423,43 @@ type PanelData struct {
 	UpdatedAt       pgtype.Timestamptz `db:"updated_at" json:"updated_at"`
 }
 
+type PodcastFeeds struct {
+	ID               pgtype.UUID        `db:"id" json:"id"`
+	LibraryID        pgtype.UUID        `db:"library_id" json:"library_id"`
+	FeedUrl          string             `db:"feed_url" json:"feed_url"`
+	PodcastGuid      pgtype.Text        `db:"podcast_guid" json:"podcast_guid"`
+	Title            string             `db:"title" json:"title"`
+	Description      pgtype.Text        `db:"description" json:"description"`
+	Host             pgtype.Text        `db:"host" json:"host"`
+	Publisher        pgtype.Text        `db:"publisher" json:"publisher"`
+	OwnerName        pgtype.Text        `db:"owner_name" json:"owner_name"`
+	OwnerEmail       pgtype.Text        `db:"owner_email" json:"owner_email"`
+	Language         pgtype.Text        `db:"language" json:"language"`
+	Tags             []string           `db:"tags" json:"tags"`
+	Explicit         bool               `db:"explicit" json:"explicit"`
+	ArtworkPath      pgtype.Text        `db:"artwork_path" json:"artwork_path"`
+	FeedType         pgtype.Text        `db:"feed_type" json:"feed_type"`
+	Medium           pgtype.Text        `db:"medium" json:"medium"`
+	IsComplete       bool               `db:"is_complete" json:"is_complete"`
+	Etag             pgtype.Text        `db:"etag" json:"etag"`
+	LastModified     pgtype.Text        `db:"last_modified" json:"last_modified"`
+	LastFetchedAt    pgtype.Timestamptz `db:"last_fetched_at" json:"last_fetched_at"`
+	LastBuildDate    pgtype.Timestamptz `db:"last_build_date" json:"last_build_date"`
+	CreatedByAdminID pgtype.UUID        `db:"created_by_admin_id" json:"created_by_admin_id"`
+	CreatedAt        pgtype.Timestamptz `db:"created_at" json:"created_at"`
+	UpdatedAt        pgtype.Timestamptz `db:"updated_at" json:"updated_at"`
+}
+
+type PodcastSubscriptions struct {
+	ID                pgtype.UUID        `db:"id" json:"id"`
+	UserID            pgtype.UUID        `db:"user_id" json:"user_id"`
+	FeedID            pgtype.UUID        `db:"feed_id" json:"feed_id"`
+	DisplayName       pgtype.Text        `db:"display_name" json:"display_name"`
+	AutoDownload      string             `db:"auto_download" json:"auto_download"`
+	AutoDownloadCount pgtype.Int4        `db:"auto_download_count" json:"auto_download_count"`
+	CreatedAt         pgtype.Timestamptz `db:"created_at" json:"created_at"`
+}
+
 // Tracks media that failed processing in a library: item-backed issues (library-type mismatches, metadata extraction failures) and row-less import failures (no media_items row exists; keyed by library + file_path). Auto-resolved by later successful scans.
 type ProcessingIssues struct {
 	ID               pgtype.UUID        `db:"id" json:"id"`
@@ -599,4 +654,5 @@ type Users struct {
 	UpdatedAt                pgtype.Timestamptz `db:"updated_at" json:"updated_at"`
 	DefaultAudioSpeed        float32            `db:"default_audio_speed" json:"default_audio_speed"`
 	KeepScreenAwake          bool               `db:"keep_screen_awake" json:"keep_screen_awake"`
+	AudioResumeRewindSeconds int32              `db:"audio_resume_rewind_seconds" json:"audio_resume_rewind_seconds"`
 }

internal/database/querier.go

@@ -1,6 +1,6 @@
 // Code generated by sqlc. DO NOT EDIT.
 // versions:
-//   sqlc v1.31.1
+//   sqlc v1.31.0
 
 package database
 
@@ -22,6 +22,8 @@ type Querier interface {
 	// ============================================
 	AddBookToKoboShelf(ctx context.Context, arg AddBookToKoboShelfParams) (KoboShelves, error)
 	// Library Folders queries
+	// purpose: 'scan' (FS-walker root, default) | 'podcast' (download root;
+	// scanner-skipped, written by the podcast download worker instead).
 	AddLibraryFolder(ctx context.Context, arg AddLibraryFolderParams) (LibraryFolders, error)
 	// Second consecutive missing scan: hide the item from all browsing while
 	// preserving reading history in case the file returns.
@@ -30,6 +32,11 @@ type Querier interface {
 	BulkUpdateFormatGroups(ctx context.Context) error
 	BulkUpdateProgressFromSync(ctx context.Context, arg BulkUpdateProgressFromSyncParams) ([]interface{}, error)
 	CheckForProgressConflicts(ctx context.Context, arg CheckForProgressConflictsParams) (int64, error)
+	// Atomic claim (§0 item 4): converts queued -> downloading in ONE
+	// statement. Two claimants racing on the same row: exactly one UPDATE
+	// matches download_state='queued'; the loser gets no rows (ErrNoRows).
+	// This is the SKIP-LOCKED guarantee without a second round-trip.
+	ClaimQueuedPodcastEpisode(ctx context.Context, id pgtype.UUID) (MediaItems, error)
 	// Cleanup expired OPDS tokens
 	CleanupExpiredOpdsTokens(ctx context.Context) error
 	CleanupExpiredRefreshTokens(ctx context.Context, dollar_1 float64) error
@@ -53,11 +60,19 @@ type Querier interface {
 	// collections page); same live filter as GetCollectionItems, plus the
 	// future-hybrid excluded exclusion and an optional library scope.
 	CountCollectionLiveItems(ctx context.Context, arg CountCollectionLiveItemsParams) (int32, error)
+	CountDownloadedEpisodes(ctx context.Context, sourceFeedID pgtype.UUID) (int64, error)
 	// Sibling-book count for sidecar naming (MetadataWriter): a dir-scoped
 	// metadata.opf may only be written when NO other non-archived item
 	// shares the directory; otherwise the sidecar goes to
 	// <basename>.opf so one book's metadata can never poison dir-mates.
 	CountOtherMediaItemsInDir(ctx context.Context, arg CountOtherMediaItemsInDirParams) (int32, error)
+	CountPodcastEpisodesForFeed(ctx context.Context, sourceFeedID pgtype.UUID) (int64, error)
+	CountPodcastFeedSubscribers(ctx context.Context, feedID pgtype.UUID) (int64, error)
+	// Sidebar/page gate (D13): >=1 podcasts-type library the user can see
+	// (COALESCE(is_visible,true) semantics — mirrors GetUserVisibleLibraries).
+	CountPodcastLibrariesForUser(ctx context.Context, userID pgtype.UUID) (int64, error)
+	CountQueuedEpisodes(ctx context.Context) (int64, error)
+	CountUndownloadedEpisodes(ctx context.Context, sourceFeedID pgtype.UUID) (int64, error)
 	// Count unlinked books for a device (ignored rows stay hidden)
 	CountUnlinkedBooks(ctx context.Context, deviceID pgtype.UUID) (int64, error)
 	// Banner count: unresolved, non-ignored queue rows across the caller's own
@@ -116,6 +131,22 @@ type Querier interface {
 	// ============================================
 	CreateOrUpdateKoboEntitlement(ctx context.Context, arg CreateOrUpdateKoboEntitlementParams) (KoboEntitlements, error)
 	// ============================================================================
+	// PODCASTS / RSS FEEDS (plan ~/.opencode/plan/podcast-integration.md §6/§8)
+	// Visibility is RLS-enforced (schema policy block) — NO query here carries
+	// visibility clauses. Feed reads run under the acting user's identity
+	// (subscriber-or-admin policy); the fetcher/worker run as first-admin.
+	// The fetcher/download worker is the SOLE writer of feed/episode metadata;
+	// users write only podcast_subscriptions rows.
+	// ============================================================================
+	// Add-feed upsert keyed on the per-library UNIQUE(library_id, feed_url):
+	// a second subscriber adding the same URL attaches to the existing row
+	// instead of erroring. DO UPDATE (not DO NOTHING) so the RETURNING works
+	// even when the conflicting row is invisible to the actor under RLS
+	// (non-admin creators cannot SELECT other users' feeds). Metadata columns
+	// are filled by the fetcher right after; this only establishes identity.
+	CreatePodcastFeed(ctx context.Context, arg CreatePodcastFeedParams) (PodcastFeeds, error)
+	CreatePodcastLibraryFolder(ctx context.Context, arg CreatePodcastLibraryFolderParams) (LibraryFolders, error)
+	// ============================================================================
 	// PROCESSING ISSUES QUERIES
 	// ============================================================================
 	// Widened upsert (faceted plan Phase 11): media_item_id and file_path
@@ -160,6 +191,11 @@ type Querier interface {
 	DeleteMediaItemFormat(ctx context.Context, id pgtype.UUID) error
 	DeleteMediaRating(ctx context.Context, arg DeleteMediaRatingParams) error
 	DeleteMediaReview(ctx context.Context, arg DeleteMediaReviewParams) error
+	// Admin-only removal (handler checks role); episodes cascade via
+	// media_items.source_feed_id ON DELETE CASCADE.
+	DeletePodcastFeed(ctx context.Context, id pgtype.UUID) (int64, error)
+	// Unsubscribe NEVER deletes feed/episode/file state (D6).
+	DeletePodcastSubscription(ctx context.Context, arg DeletePodcastSubscriptionParams) (int64, error)
 	DeleteProcessingIssue(ctx context.Context, id pgtype.UUID) (ProcessingIssues, error)
 	DeleteReadingProgress(ctx context.Context, arg DeleteReadingProgressParams) error
 	DeleteSavedFilter(ctx context.Context, arg DeleteSavedFilterParams) (SavedFilters, error)
@@ -170,6 +206,7 @@ type Querier interface {
 	// Delete unlinked book
 	DeleteUnlinkedBook(ctx context.Context, id pgtype.UUID) error
 	DeleteUser(ctx context.Context, id pgtype.UUID) error
+	DeleteUserSettingByKey(ctx context.Context, arg DeleteUserSettingByKeyParams) error
 	DeleteUserSystemCollection(ctx context.Context, arg DeleteUserSystemCollectionParams) error
 	// Device deletion (deregister on logout): detach the device's reading
 	// history instead of deleting it — these rows are real captured stats
@@ -225,11 +262,25 @@ type Querier interface {
 	// rename can't spawn duplicates on the next scan.
 	GetCollectionByImportedSeries(ctx context.Context, arg GetCollectionByImportedSeriesParams) (Collections, error)
 	GetCollectionByUserAndName(ctx context.Context, arg GetCollectionByUserAndNameParams) (Collections, error)
+	// The collection queue's downloaded episodes in collection order
+	// (plan item 2.3 "Play collection as sequence"): position-first, then
+	// added_at. INNER JOIN — the chain steps only through episodes the
+	// requester can see (RLS drops unsubscribed feeds) and that are
+	// downloaded (invariant 6: un-downloaded episodes are never tracked).
+	GetCollectionEpisodesOrdered(ctx context.Context, collectionID pgtype.UUID) ([]GetCollectionEpisodesOrderedRow, error)
 	// Get collection items
 	// A2c: position-first ordering (playlist reading order); un-positioned
 	// legacy items keep added_at ASC after the positioned tail — NOT DESC,
 	// which walked a playlist backwards.
 	GetCollectionItems(ctx context.Context, collectionID pgtype.UUID) ([]GetCollectionItemsRow, error)
+	// Episode-collections detail (plan item 2.3): the collection's full
+	// membership for the SSR detail page, with per-row visibility. LEFT
+	// JOIN + RLS: rows whose media is invisible to the requester
+	// (unsubscribed episode — or a family-lens hidden book) survive with
+	// NULL media columns and visible=false, so the page can render an
+	// "unavailable" hint instead of silently vanishing them. Rows visible
+	// but archived/missing-scan keep the old INNER-JOIN behavior (dropped).
+	GetCollectionItemsDetailed(ctx context.Context, collectionID pgtype.UUID) ([]GetCollectionItemsDetailedRow, error)
 	// The excluded filter lives here (was: callers filtered ci.excluded in
 	// Go over a 69-field row + a 45-field converter to MediaItems; embed
 	// returns the real row so neither exists).
@@ -421,6 +472,25 @@ type Querier interface {
 	// Get OPDS tokens by device
 	GetOpdsTokensByDevice(ctx context.Context, deviceID pgtype.UUID) ([]OpdsTokens, error)
 	GetPanelData(ctx context.Context, arg GetPanelDataParams) (PanelData, error)
+	// Continue aggregation (plan 2.5): the caller's IN-PROGRESS episodes
+	// (0 < pct < finishedThreshold) across ALL subscribed feeds, most
+	// recent first. Go groups per feed — the first row per feed IS the
+	// continue target.
+	GetPodcastContinueForUser(ctx context.Context, userID pgtype.UUID) ([]GetPodcastContinueForUserRow, error)
+	GetPodcastEpisode(ctx context.Context, id pgtype.UUID) (MediaItems, error)
+	// Per-user progress for a feed's episodes (row badges + the Phase-2
+	// continue fields). reading_progress is UNIQUE per (media_item, user).
+	GetPodcastEpisodeProgress(ctx context.Context, arg GetPodcastEpisodeProgressParams) ([]GetPodcastEpisodeProgressRow, error)
+	GetPodcastFeed(ctx context.Context, id pgtype.UUID) (PodcastFeeds, error)
+	// URL-migration dedupe: same canonical guid within the library reuses
+	// the existing row.
+	GetPodcastFeedByGuid(ctx context.Context, arg GetPodcastFeedByGuidParams) (PodcastFeeds, error)
+	// The download root (purpose='podcast') of a podcasts library.
+	GetPodcastLibraryFolder(ctx context.Context, libraryID pgtype.UUID) ([]LibraryFolders, error)
+	GetPodcastSubscription(ctx context.Context, arg GetPodcastSubscriptionParams) (PodcastSubscriptions, error)
+	// The podcasts library/ies visible to the user (add-feed target list;
+	// a self-hosted instance typically has exactly one).
+	GetPodcastsLibraryForUser(ctx context.Context, userID pgtype.UUID) ([]GetPodcastsLibraryForUserRow, error)
 	// Get most popular books for a user
 	GetPopularBooks(ctx context.Context, arg GetPopularBooksParams) ([]GetPopularBooksRow, error)
 	// Single-issue read for the resolve/delete endpoints' family-lens gate.
@@ -521,14 +591,14 @@ type Querier interface {
 	// END instant (the push time; NULL falls back to created_at) — the same
 	// value the capture stamps every write with.
 	GetUserReadingSessionsRange(ctx context.Context, arg GetUserReadingSessionsRangeParams) ([]GetUserReadingSessionsRangeRow, error)
+	// ============================================================================
+	// GENERIC reader_settings KV (podcast order override, plan D5: key
+	// podcast_order:<feed_id>). The pre-existing reader_settings queries are
+	// hard-keyed to 'reader_settings'; these two are the generic form.
+	// ============================================================================
+	GetUserSettingByKey(ctx context.Context, arg GetUserSettingByKeyParams) ([]byte, error)
 	GetUserVisibleLibraries(ctx context.Context, userID pgtype.UUID) ([]GetUserVisibleLibrariesRow, error)
 	GetVisibleLibraryMediaCounts(ctx context.Context, userID pgtype.UUID) ([]GetVisibleLibraryMediaCountsRow, error)
-	// Smart ("dynamic") collections: rules are evaluated LIVE at render, so
-	// membership comes from the candidate book set instead of collection_items.
-	// The set is scoped to the requesting user's visible libraries (family
-	// lens enforced at the source) and live items only — the same filters the
-	// static collection paths apply.
-	GetVisibleLiveMediaForRules(ctx context.Context, arg GetVisibleLiveMediaForRulesParams) ([]MediaItems, error)
 	HasRecentConflictResolution(ctx context.Context, arg HasRecentConflictResolutionParams) (bool, error)
 	// Soft-dismiss one row (ownership is checked by the caller against the
 	// device's user_id before calling this).
@@ -540,6 +610,12 @@ type Querier interface {
 	// Link unlinked book to media item
 	LinkUnlinkedBook(ctx context.Context, arg LinkUnlinkedBookParams) (UnlinkedBooks, error)
 	ListAllConflictsByUserAndStatus(ctx context.Context, arg ListAllConflictsByUserAndStatusParams) ([]ListAllConflictsByUserAndStatusRow, error)
+	// Admin all-podcasts view (§0 item 8 / Phase-4 item 1): EVERY feed from
+	// EVERY user with its maintenance aggregates. Runs under the first-
+	// admin identity (invariant 12 — same contract as the poller). Sorted
+	// by disk usage DESC: the maintenance question is "what's eating the
+	// disk", so the hogs surface first.
+	ListAllPodcastFeedsAdmin(ctx context.Context) ([]ListAllPodcastFeedsAdminRow, error)
 	ListAllSyncQueueItems(ctx context.Context, arg ListAllSyncQueueItemsParams) ([]ListAllSyncQueueItemsRow, error)
 	ListConflictsByUser(ctx context.Context, userID pgtype.UUID) ([]ListConflictsByUserRow, error)
 	// ============================================
@@ -552,6 +628,9 @@ type Querier interface {
 	ListDeletedAnnotationsForBook(ctx context.Context, arg ListDeletedAnnotationsForBookParams) ([]ListDeletedAnnotationsForBookRow, error)
 	ListDevicesByType(ctx context.Context, deviceType string) ([]Devices, error)
 	ListDevicesByUser(ctx context.Context, userID pgtype.UUID) ([]Devices, error)
+	// Download-worker policy union (D6): most aggressive policy wins for
+	// what arrives. Runs under an admin identity (background job).
+	ListFeedSubscribers(ctx context.Context, feedID pgtype.UUID) ([]ListFeedSubscribersRow, error)
 	// Items hidden from libraries: either missing from disk (waiting on the
 	// second scan that archives them) or already archived and pending purge.
 	// Feeds the admin archived-items page. match_* columns identify an active
@@ -563,6 +642,9 @@ type Querier interface {
 	// collapsible "Ignored" section so a misclick is recoverable.
 	ListIgnoredUnlinkedBooksForUser(ctx context.Context, userID pgtype.UUID) ([]ListIgnoredUnlinkedBooksForUserRow, error)
 	ListLibraries(ctx context.Context) ([]ListLibrariesRow, error)
+	// Podcasts exclusion (plan §9): generic book surfaces never list
+	// podcast episodes (library_type_name is the denormalized, trigger/
+	// fetcher-maintained discriminator; NULL rows predate it and stay).
 	ListMediaItems(ctx context.Context, arg ListMediaItemsParams) ([]ListMediaItemsRow, error)
 	ListMediaItemsByLibrary(ctx context.Context, libraryID pgtype.UUID) ([]ListMediaItemsByLibraryRow, error)
 	ListMediaItemsByLibraryIncludingArchived(ctx context.Context, libraryID pgtype.UUID) ([]ListMediaItemsByLibraryIncludingArchivedRow, error)
@@ -571,18 +653,35 @@ type Querier interface {
 	// List media items that have no stored SHA-256 (imported before hashing existed)
 	ListMediaItemsMissingHash(ctx context.Context) ([]MediaItems, error)
 	ListMediaItemsSorted(ctx context.Context, arg ListMediaItemsSortedParams) ([]ListMediaItemsSortedRow, error)
+	// The podcasts page: the acting user's subscriptions with feed metadata
+	// and the user-owned subscription fields. Facet values derive in Go.
+	ListMyPodcastFeeds(ctx context.Context, userID pgtype.UUID) ([]ListMyPodcastFeedsRow, error)
 	ListPendingHashConflicts(ctx context.Context) ([]ListPendingHashConflictsRow, error)
 	// DeviceID NULL = every device's pending items (the queue processor's
 	// enumerator passes NULL; a bare `device_id = NULL` never matches, which
 	// silently disabled the offline flush until 2026-10-17).
 	ListPendingSyncQueueItems(ctx context.Context, arg ListPendingSyncQueueItemsParams) ([]SyncQueue, error)
+	// Feed detail / sequence source. RLS scopes to subscriber-or-admin;
+	// ordering: native position, then pub date, then guid (stable across
+	// re-fetches for historical rows whose position went stale).
+	ListPodcastEpisodes(ctx context.Context, sourceFeedID pgtype.UUID) ([]MediaItems, error)
+	// Poller worklist (admin identity): every feed with its validators and
+	// the back-off flag (is_complete feeds poll rarely, never disappear).
+	ListPodcastFeedsForPolling(ctx context.Context) ([]PodcastFeeds, error)
 	// LEFT JOIN (Phase 11): row-less import failures have NO media_items
 	// row — an inner join would silently drop the ledger's newest kind.
 	// Item-side projections COALESCE to keep them non-nullable strings;
 	// the display file path falls back to the issue's own for row-less rows.
 	ListProcessingIssuesByLibrary(ctx context.Context, libraryID pgtype.UUID) ([]ListProcessingIssuesByLibraryRow, error)
+	// Fair-share candidate scan (§0 item 4): the download pool claims in
+	// TICKET order (download_queue_pos = SEQUENCE nextval at enqueue — the
+	// enqueue loop's ordering IS the download order, item 5). The worker
+	// groups these per feed and prefers the least-recently-served feed.
+	ListQueuedPodcastEpisodes(ctx context.Context) ([]ListQueuedPodcastEpisodesRow, error)
 	ListSyncConflictsByMediaItem(ctx context.Context, arg ListSyncConflictsByMediaItemParams) ([]SyncConflicts, error)
 	ListSyncConflictsByUser(ctx context.Context, userID pgtype.UUID) ([]ListSyncConflictsByUserRow, error)
+	// Download All + policy union worklist.
+	ListUndownloadedEpisodes(ctx context.Context, sourceFeedID pgtype.UUID) ([]MediaItems, error)
 	// List unresolved unlinked books with pagination
 	ListUnresolvedUnlinkedBooks(ctx context.Context, arg ListUnresolvedUnlinkedBooksParams) ([]ListUnresolvedUnlinkedBooksRow, error)
 	// Page list: unresolved queue rows for the caller's own devices (ignored
@@ -591,6 +690,17 @@ type Querier interface {
 	ListUsers(ctx context.Context) ([]ListUsersRow, error)
 	// First consecutive scan that cannot find the file on disk.
 	MarkMediaItemMissing(ctx context.Context, id pgtype.UUID) error
+	// SetPodcastEpisodeDownloading RETIRED (§0 item 4): the pair
+	// GetNextQueuedEpisode + SetPodcastEpisodeDownloading raced under
+	// concurrency; ClaimQueuedPodcastEpisode does both atomically.
+	// SUCCESS-only write (invariant 7): the file is on disk and probed.
+	// ffprobe facts (real duration via audio_metadata, precise size/mime)
+	// reconcile the feed's declared values.
+	MarkPodcastEpisodeDownloaded(ctx context.Context, arg MarkPodcastEpisodeDownloadedParams) (MediaItems, error)
+	// Failure/rejection write: state + reason + whether a human already
+	// intervened (manual_retry resets the auto-retry budget). Existing
+	// metadata/file facts are never cleared.
+	MarkPodcastEpisodeFailed(ctx context.Context, arg MarkPodcastEpisodeFailedParams) error
 	// Content reappeared at a new location while the old path vanished: treat as
 	// a move. Repoints the row (and refreshes file-level fields) so reading
 	// history follows the book; archive/missing state is cleared separately.
@@ -625,6 +735,17 @@ type Querier interface {
 	ReorderCollectionItems(ctx context.Context, arg ReorderCollectionItemsParams) error
 	// Re-parent all child rows of p_source onto p_target (defined in schema.sql)
 	ReparentMediaItemChildren(ctx context.Context, arg ReparentMediaItemChildrenParams) error
+	// Admin file deletion (§0 item 8): ONE downloaded episode's disk file
+	// was removed — reset the row to pending and clear the download-derived
+	// facts. file_path is deliberately KEPT: it is the fetch-maintained
+	// layout (written at fetch time, not download time), so a re-download
+	// lands on the same path. The poll upsert never touches download_state,
+	// so a deleted file stays deleted (invariant 2: polling never deletes;
+	// admins delete knowingly).
+	ResetPodcastEpisodeFileDeleted(ctx context.Context, id pgtype.UUID) error
+	// Same reset for EVERY downloaded episode of a feed (per-feed
+	// "Delete files" action on the admin view).
+	ResetPodcastFeedFilesDeleted(ctx context.Context, sourceFeedID pgtype.UUID) (int64, error)
 	ResetSystemCollectionMetadata(ctx context.Context, arg ResetSystemCollectionMetadataParams) error
 	ResolveHashConflict(ctx context.Context, arg ResolveHashConflictParams) error
 	// Ledger self-heal (faceted plan Phase 11): a scan that processes the
@@ -649,19 +770,44 @@ type Querier interface {
 	RevokeOpdsToken(ctx context.Context, token string) error
 	RevokeRefreshToken(ctx context.Context, token pgtype.UUID) error
 	SearchAuthorValues(ctx context.Context, arg SearchAuthorValuesParams) ([]SearchAuthorValuesRow, error)
+	// Episode picker scope (plan item 2.3): YOUR subscriptions' episodes.
+	// The podcasts exclusion in SearchMediaItemsUnified stays (books
+	// surfaces never list podcasts — plan §9); THIS query is the episode
+	// picker's dedicated scope. Visibility is the engine's job (D20): the
+	// media_items policy drops episodes from unsubscribed feeds and the
+	// podcast_feeds join is governed the same way — non-subscribers see
+	// nothing here without any Go-side clause.
+	SearchEpisodesForPicker(ctx context.Context, q_ string) ([]SearchEpisodesForPickerRow, error)
 	SearchGenreValues(ctx context.Context, arg SearchGenreValuesParams) ([]SearchGenreValuesRow, error)
 	SearchLanguageValues(ctx context.Context, arg SearchLanguageValuesParams) ([]SearchLanguageValuesRow, error)
 	// Search Media Items queries
 	SearchMediaItems(ctx context.Context, arg SearchMediaItemsParams) ([]SearchMediaItemsRow, error)
 	SearchMediaItemsUnified(ctx context.Context, arg SearchMediaItemsUnifiedParams) ([]SearchMediaItemsUnifiedRow, error)
+	// Subscription-scoped episode search (D14): title + description cover
+	// un-downloaded AND historical rows (fetcher persists metadata-only).
+	// RLS already restricts to the caller's subscriptions — the explicit
+	// source filter keeps scan items out of podcast search.
+	SearchMyPodcastEpisodes(ctx context.Context, searchQuery string) ([]MediaItems, error)
+	// Feed-title AND subscription-display-name search (D14) + the facet
+	// fields the podcasts page filters on.
+	SearchMyPodcastFeeds(ctx context.Context, arg SearchMyPodcastFeedsParams) ([]SearchMyPodcastFeedsRow, error)
 	SearchNarratorValues(ctx context.Context, arg SearchNarratorValuesParams) ([]SearchNarratorValuesRow, error)
 	SearchSeriesValues(ctx context.Context, arg SearchSeriesValuesParams) ([]SearchSeriesValuesRow, error)
 	SearchTagsValues(ctx context.Context, arg SearchTagsValuesParams) ([]SearchTagsValuesRow, error)
 	SetDeviceSyncEnabled(ctx context.Context, arg SetDeviceSyncEnabledParams) (Devices, error)
 	// Library Visibility queries
 	SetLibraryVisibility(ctx context.Context, arg SetLibraryVisibilityParams) (LibraryVisibility, error)
+	// Enqueue: pending|failed -> queued, stamped with the NEXT QUEUE TICKET
+	// (§0 items 4/5) so claim order = enqueue order (the effective
+	// listening order the enqueue loop applied). Attempts increment at
+	// enqueue so a crashed worker still counts the try.
+	SetPodcastEpisodeDownloadQueued(ctx context.Context, id pgtype.UUID) (int64, error)
+	// Manual retry (same endpoint as download): reset to pending so the
+	// worker re-claims it; the reason text clears with the fresh attempt.
+	SetPodcastEpisodePending(ctx context.Context, id pgtype.UUID) error
 	// Set system config
 	SetSystemConfig(ctx context.Context, arg SetSystemConfigParams) (SystemConfig, error)
+	SubscribePodcastFeed(ctx context.Context, arg SubscribePodcastFeedParams) (PodcastSubscriptions, error)
 	SyncLibraryTypeExtensions(ctx context.Context, arg SyncLibraryTypeExtensionsParams) error
 	TombstoneMediaBookmarkByDedupKey(ctx context.Context, arg TombstoneMediaBookmarkByDedupKeyParams) error
 	TombstoneMediaBookmarkByID(ctx context.Context, arg TombstoneMediaBookmarkByIDParams) error
@@ -674,11 +820,12 @@ type Querier interface {
 	UnignoreUnlinkedBook(ctx context.Context, id pgtype.UUID) error
 	// Update collection
 	UpdateCollection(ctx context.Context, arg UpdateCollectionParams) (Collections, error)
-	// Replace ONLY the auto_assign_rules document (rule CRUD: read-modify-write
-	// of the JSONB without touching metadata stale clients did not send).
-	UpdateCollectionRulesJSON(ctx context.Context, arg UpdateCollectionRulesJSONParams) (Collections, error)
 	// A2c §2.4: provenance + playlist flags on an existing collection without
 	// touching the rest of the row.
+	// (Historical bug, fixed 2026-10-30: this wrote `updated_at = NOW()` but
+	// collections has no updated_at column — the scanner's series-import
+	// call has been failing silently, its error discarded. Fixed by
+	// dropping the phantom column write.)
 	UpdateCollectionViewSettings(ctx context.Context, arg UpdateCollectionViewSettingsParams) error
 	UpdateDashboardPreferences(ctx context.Context, arg UpdateDashboardPreferencesParams) (UserDashboardPreferences, error)
 	UpdateDevice(ctx context.Context, arg UpdateDeviceParams) (Devices, error)
@@ -726,6 +873,20 @@ type Querier interface {
 	UpdateMediaNoteForSync(ctx context.Context, arg UpdateMediaNoteForSyncParams) (MediaNotes, error)
 	UpdateMediaRating(ctx context.Context, arg UpdateMediaRatingParams) (MediaRatings, error)
 	UpdatePassword(ctx context.Context, arg UpdatePasswordParams) error
+	UpdatePodcastFeedArtwork(ctx context.Context, arg UpdatePodcastFeedArtworkParams) error
+	// Post-fetch bookkeeping: validators for conditional GETs, the fetch
+	// timestamp, and the feed's own last-build marker.
+	UpdatePodcastFeedFetch(ctx context.Context, arg UpdatePodcastFeedFetchParams) (PodcastFeeds, error)
+	// Fetcher-only full metadata refresh (parse result). podcast_guid is
+	// set once a feed declares it; new_feed_url handling is a dedicated
+	// query below.
+	UpdatePodcastFeedMetadata(ctx context.Context, arg UpdatePodcastFeedMetadataParams) (PodcastFeeds, error)
+	// itunes:new-feed-url (or a permanent HTTP redirect chain): the feed has
+	// MOVED. Rewrite feed_url so future polls hit the new location; the row
+	// (and every subscription/episode) stays put.
+	UpdatePodcastFeedRedirect(ctx context.Context, arg UpdatePodcastFeedRedirectParams) error
+	// User-owned fields only (D9): display name + auto-download policy.
+	UpdatePodcastSubscription(ctx context.Context, arg UpdatePodcastSubscriptionParams) (PodcastSubscriptions, error)
 	UpdateReadingProgress(ctx context.Context, arg UpdateReadingProgressParams) (ReadingProgress, error)
 	UpdateReadingSpeed(ctx context.Context, arg UpdateReadingSpeedParams) (ReadingSpeed, error)
 	UpdateSavedFilter(ctx context.Context, arg UpdateSavedFilterParams) (SavedFilters, error)
@@ -733,6 +894,7 @@ type Querier interface {
 	UpdateSystemSetting(ctx context.Context, arg UpdateSystemSettingParams) error
 	// Update universal progress
 	UpdateUniversalProgress(ctx context.Context, arg UpdateUniversalProgressParams) (ReadingProgress, error)
+	UpdateUserAudioResumeRewind(ctx context.Context, arg UpdateUserAudioResumeRewindParams) error
 	UpdateUserDefaultAudioSpeed(ctx context.Context, arg UpdateUserDefaultAudioSpeedParams) error
 	UpdateUserKeepScreenAwake(ctx context.Context, arg UpdateUserKeepScreenAwakeParams) error
 	UpdateUserMaxDevices(ctx context.Context, arg UpdateUserMaxDevicesParams) (Users, error)
@@ -751,11 +913,23 @@ type Querier interface {
 	// content identical), and after a manifest-gated re-extract.
 	UpsertMediaScanState(ctx context.Context, arg UpsertMediaScanStateParams) error
 	UpsertPanelData(ctx context.Context, arg UpsertPanelDataParams) (PanelData, error)
+	// ============================================================================
+	// Podcast episodes — media_items rows with source='podcast_episode'
+	// ============================================================================
+	// Fetcher-only upsert keyed on the partial unique index
+	// (source_feed_id, episode_guid). DO UPDATE refreshes FETCH-OWNED
+	// metadata only. Write-after-success guard (AGENTS.md gotcha #7):
+	// download state/error/attempts are NEVER touched here, and the
+	// downloaded file's facts (path/size/mime/cover) survive renumbering —
+	// only un-downloaded rows follow the re-derived file_path.
+	// authors_search mirrors authors (trigram search column).
+	UpsertPodcastEpisode(ctx context.Context, arg UpsertPodcastEpisodeParams) (MediaItems, error)
 	UpsertReaderSettings(ctx context.Context, arg UpsertReaderSettingsParams) (ReaderSettings, error)
 	UpsertSystemSetting(ctx context.Context, arg UpsertSystemSettingParams) (SystemSettings, error)
 	// Ingest: insert or refresh the queue row for a device file that could not
 	// be identified. One row per (device, content_id); re-syncs bump last_seen.
 	UpsertUnlinkedBook(ctx context.Context, arg UpsertUnlinkedBookParams) (UnlinkedBooks, error)
+	UpsertUserSettingByKey(ctx context.Context, arg UpsertUserSettingByKeyParams) error
 }
 
 var _ Querier = (*Queries)(nil)

internal/database/queries/queries.sql

@@ -30,6 +30,7 @@ SELECT
     u.timezone,
     u.default_audio_speed,
     u.keep_screen_awake,
+    u.audio_resume_rewind_seconds,
     (SELECT COUNT(*) FROM devices WHERE user_id = u.id) as device_count
 FROM users u
 WHERE u.id = $1;
@@ -97,7 +98,9 @@ DELETE FROM libraries WHERE id = $1;
 
 -- Library Folders queries
 -- name: AddLibraryFolder :one
-INSERT INTO library_folders (library_id, folder_path) VALUES ($1, $2) RETURNING *;
+-- purpose: 'scan' (FS-walker root, default) | 'podcast' (download root;
+-- scanner-skipped, written by the podcast download worker instead).
+INSERT INTO library_folders (library_id, folder_path, purpose) VALUES ($1, $2, $3) RETURNING *;
 
 -- name: GetLibraryFolders :many
 SELECT * FROM library_folders WHERE library_id = $1 ORDER BY created_at;
@@ -178,12 +181,16 @@ SELECT * FROM media_items WHERE library_id = $1 AND file_path = $2;
 SELECT * FROM media_items WHERE id = $1;
 
 -- name: ListMediaItems :many
+-- Podcasts exclusion (plan §9): generic book surfaces never list
+-- podcast episodes (library_type_name is the denormalized, trigger/
+-- fetcher-maintained discriminator; NULL rows predate it and stay).
 SELECT mi.*, l.name as library_name, lt.name as library_type_name
 FROM media_items mi
 JOIN libraries l ON mi.library_id = l.id
 JOIN library_types lt ON l.library_type_id = lt.id
 WHERE mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts')
   AND (sqlc.narg('visible_library_ids')::uuid[] IS NULL OR mi.library_id = ANY(sqlc.narg('visible_library_ids')::uuid[]))
 ORDER BY mi.created_at DESC LIMIT $1 OFFSET $2;
 
@@ -212,6 +219,7 @@ JOIN libraries l ON mi.library_id = l.id
 JOIN library_types lt ON l.library_type_id = lt.id
 WHERE mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.library_id = sqlc.narg('library_id')
 ORDER BY
     CASE
@@ -519,6 +527,9 @@ UPDATE users SET default_audio_speed = $2, updated_at = NOW() WHERE id = $1;
 -- name: UpdateUserKeepScreenAwake :exec
 UPDATE users SET keep_screen_awake = $2, updated_at = NOW() WHERE id = $1;
 
+-- name: UpdateUserAudioResumeRewind :exec
+UPDATE users SET audio_resume_rewind_seconds = $2, updated_at = NOW() WHERE id = $1;
+
 -- Per-book playback speed preference (player plan §8).
 
 -- name: GetAudioPlaybackSpeed :one
@@ -658,6 +669,7 @@ LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.na
  WHERE COALESCE(lv.is_visible, true) = true
    AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
    AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
    AND (
        mi.title ILIKE sqlc.narg('search_pattern') OR
@@ -748,6 +760,7 @@ LEFT JOIN reading_progress rp ON rp.media_item_id = mi.id AND rp.user_id = sqlc.
 WHERE COALESCE(lv.is_visible, true) = true
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND (sqlc.narg('library_id')::uuid IS NULL
   OR mi.library_id = sqlc.narg('library_id')::uuid)
   -- Fuzzy author filter — per-element against the authors[] array (A2c
@@ -923,6 +936,26 @@ ORDER BY
   mi.title ASC
 LIMIT sqlc.narg('limit') OFFSET sqlc.narg('offset');
 
+-- name: SearchEpisodesForPicker :many
+-- Episode picker scope (plan item 2.3): YOUR subscriptions' episodes.
+-- The podcasts exclusion in SearchMediaItemsUnified stays (books
+-- surfaces never list podcasts — plan §9); THIS query is the episode
+-- picker's dedicated scope. Visibility is the engine's job (D20): the
+-- media_items policy drops episodes from unsubscribed feeds and the
+-- podcast_feeds join is governed the same way — non-subscribers see
+-- nothing here without any Go-side clause.
+SELECT mi.id, mi.title, mi.author, mi.cover_image_path, mi.library_id,
+       mi.format_group, mi.download_state,
+       pf.title AS feed_title
+FROM media_items mi
+JOIN podcast_feeds pf ON mi.source_feed_id = pf.id
+WHERE mi.source = 'podcast_episode'
+  AND mi.archived_at IS NULL
+  AND mi.missing_scan_count = 0
+  AND (sqlc.arg('q')::text = '' OR mi.title ILIKE '%' || sqlc.arg('q')::text || '%')
+ORDER BY pf.title ASC, mi.feed_position ASC NULLS LAST
+LIMIT 60;
+
 -- name: SearchAuthorValues :many
 SELECT
   mi.author as value,
@@ -932,6 +965,7 @@ FROM media_items mi
 JOIN libraries l ON mi.library_id = l.id
 LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.narg('user_id')
 WHERE COALESCE(lv.is_visible, true) = true
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
   AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
@@ -951,6 +985,7 @@ FROM media_items mi
 JOIN libraries l ON mi.library_id = l.id
 LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.narg('user_id')
 WHERE COALESCE(lv.is_visible, true) = true
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
   AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
@@ -971,6 +1006,7 @@ JOIN libraries l ON mi.library_id = l.id
 LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.narg('user_id')
 CROSS JOIN LATERAL unnest(mi.tags_search) AS tag
 WHERE COALESCE(lv.is_visible, true) = true
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
   AND tag IS NOT NULL
   AND word_similarity(sqlc.narg('search_query'), tag) > 0.3
@@ -988,6 +1024,7 @@ JOIN libraries l ON mi.library_id = l.id
 LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.narg('user_id')
 CROSS JOIN LATERAL unnest(mi.narrators) AS narrator
 WHERE COALESCE(lv.is_visible, true) = true
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
   AND narrator IS NOT NULL
   AND word_similarity(sqlc.narg('search_query'), narrator) > 0.3
@@ -1004,6 +1041,7 @@ FROM media_items mi
 JOIN libraries l ON mi.library_id = l.id
 LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.narg('user_id')
 WHERE COALESCE(lv.is_visible, true) = true
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
   AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
@@ -1023,6 +1061,7 @@ FROM media_items mi
 JOIN libraries l ON mi.library_id = l.id
 LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.narg('user_id')
 WHERE COALESCE(lv.is_visible, true) = true
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
   AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
@@ -1061,6 +1100,7 @@ JOIN libraries l ON mi.library_id = l.id
 LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.narg('user_id')
 CROSS JOIN LATERAL unnest(mi.authors) AS a
 WHERE COALESCE(lv.is_visible, true) = true
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
   AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
@@ -1129,6 +1169,7 @@ JOIN libraries l ON mi.library_id = l.id
 LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.narg('user_id')
 CROSS JOIN LATERAL unnest(mi.publishers) AS p
 WHERE COALESCE(lv.is_visible, true) = true
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
   AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
@@ -1196,6 +1237,7 @@ JOIN libraries l ON mi.library_id = l.id
 LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.narg('user_id')
 CROSS JOIN LATERAL unnest(mi.narrators) AS n
 WHERE COALESCE(lv.is_visible, true) = true
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
   AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
@@ -1265,6 +1307,7 @@ JOIN libraries l ON mi.library_id = l.id
 LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.narg('user_id')
 CROSS JOIN LATERAL unnest(mi.tags) AS t
 WHERE COALESCE(lv.is_visible, true) = true
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
   AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
@@ -1333,6 +1376,7 @@ FROM media_items mi
 JOIN libraries l ON mi.library_id = l.id
 LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.narg('user_id')
 WHERE COALESCE(lv.is_visible, true) = true
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
   AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
@@ -1404,6 +1448,7 @@ FROM media_items mi
 JOIN libraries l ON mi.library_id = l.id
 LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.narg('user_id')
 WHERE COALESCE(lv.is_visible, true) = true
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
   AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
@@ -1475,6 +1520,7 @@ FROM media_items mi
 JOIN libraries l ON mi.library_id = l.id
 LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.narg('user_id')
 WHERE COALESCE(lv.is_visible, true) = true
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
   AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
@@ -1546,6 +1592,7 @@ FROM media_items mi
 JOIN libraries l ON mi.library_id = l.id
 LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.narg('user_id')
 WHERE COALESCE(lv.is_visible, true) = true
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
   AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
@@ -1613,6 +1660,7 @@ FROM media_items mi
 JOIN libraries l ON mi.library_id = l.id
 LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.narg('user_id')
 WHERE COALESCE(lv.is_visible, true) = true
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
   AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
@@ -1680,6 +1728,7 @@ FROM media_items mi
 JOIN libraries l ON mi.library_id = l.id
 LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.narg('user_id')
 WHERE COALESCE(lv.is_visible, true) = true
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
   AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
@@ -1747,6 +1796,7 @@ FROM media_items mi
 JOIN libraries l ON mi.library_id = l.id
 LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.narg('user_id')
 WHERE COALESCE(lv.is_visible, true) = true
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
   AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
@@ -1818,6 +1868,7 @@ FROM media_items mi
 JOIN libraries l ON mi.library_id = l.id
 LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.narg('user_id')
 WHERE COALESCE(lv.is_visible, true) = true
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
   AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
@@ -1889,6 +1940,7 @@ FROM media_items mi
 JOIN libraries l ON mi.library_id = l.id
 LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.narg('user_id')
 WHERE COALESCE(lv.is_visible, true) = true
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
   AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
@@ -1960,6 +2012,7 @@ FROM media_items mi
 JOIN libraries l ON mi.library_id = l.id
 LEFT JOIN library_visibility lv ON l.id = lv.library_id AND lv.user_id = sqlc.narg('user_id')
 WHERE COALESCE(lv.is_visible, true) = true
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
   AND (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
@@ -2604,6 +2657,7 @@ SELECT
  FROM media_items mi
  JOIN reading_history rh ON rh.media_item_id = mi.id
  WHERE rh.user_id = $1
+   AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
  GROUP BY mi.id, mi.title, mi.author
  ORDER BY read_count DESC
  LIMIT $2;
@@ -2666,6 +2720,7 @@ SELECT
 FROM reading_history rh
 JOIN media_items mi ON mi.id = rh.media_item_id
 WHERE rh.user_id = sqlc.arg('user_id')
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND COALESCE(rh.reading_session_end, rh.created_at) >= sqlc.arg('start_time')
   AND COALESCE(rh.reading_session_end, rh.created_at) <= sqlc.arg('end_time')
   AND (sqlc.narg('visible_library_ids')::uuid[] IS NULL OR mi.library_id = ANY(sqlc.narg('visible_library_ids')::uuid[]))
@@ -3438,14 +3493,18 @@ DELETE FROM device_file_aliases WHERE id = $1;
 
 -- Create collection
 -- name: CreateCollection :one
-INSERT INTO collections (user_id, name, description, color, icon, auto_assign_rules, view_settings, show_on_dashboard)
+INSERT INTO collections (user_id, name, description, color, icon, view_settings, show_on_dashboard, filter_tree)
 VALUES ($1, $2, $3, $4, $5, $6, $7, $8)
 RETURNING *;
 
 -- A2c §2.4: provenance + playlist flags on an existing collection without
 -- touching the rest of the row.
 -- name: UpdateCollectionViewSettings :exec
-UPDATE collections SET view_settings = $2, updated_at = NOW() WHERE id = $1;
+-- (Historical bug, fixed 2026-10-30: this wrote `updated_at = NOW()` but
+-- collections has no updated_at column — the scanner's series-import
+-- call has been failing silently, its error discarded. Fixed by
+-- dropping the phantom column write.)
+UPDATE collections SET view_settings = $2 WHERE id = $1;
 
 -- Get collection
 -- name: GetCollection :one
@@ -3473,16 +3532,11 @@ SET
     description = $3,
     color = $4,
     icon = $5,
-    auto_assign_rules = $6,
+    view_settings = $6,
-    view_settings = $7
+    filter_tree = $7
 WHERE id = $1
 RETURNING *;
 
--- Replace ONLY the auto_assign_rules document (rule CRUD: read-modify-write
--- of the JSONB without touching metadata stale clients did not send).
--- name: UpdateCollectionRulesJSON :one
-UPDATE collections SET auto_assign_rules = $2 WHERE id = $1 RETURNING *;
-
 -- Delete collection
 -- name: DeleteCollection :exec
 DELETE FROM collections WHERE id = $1;
@@ -3545,6 +3599,42 @@ WHERE ci.collection_id = $1
   AND mi.missing_scan_count = 0
 ORDER BY ci.position ASC NULLS LAST, ci.added_at ASC;
 
+-- name: GetCollectionItemsDetailed :many
+-- Episode-collections detail (plan item 2.3): the collection's full
+-- membership for the SSR detail page, with per-row visibility. LEFT
+-- JOIN + RLS: rows whose media is invisible to the requester
+-- (unsubscribed episode — or a family-lens hidden book) survive with
+-- NULL media columns and visible=false, so the page can render an
+-- "unavailable" hint instead of silently vanishing them. Rows visible
+-- but archived/missing-scan keep the old INNER-JOIN behavior (dropped).
+SELECT ci.media_item_id, ci.position, ci.added_at,
+       (mi.id IS NOT NULL)::bool AS visible,
+       mi.title, mi.author, mi.cover_image_path, mi.format_group,
+       mi.source, mi.download_state,
+       pf.title AS feed_title
+FROM collection_items ci
+LEFT JOIN media_items mi ON ci.media_item_id = mi.id
+LEFT JOIN podcast_feeds pf ON mi.source_feed_id = pf.id
+WHERE ci.collection_id = $1
+  AND (mi.id IS NULL OR (mi.archived_at IS NULL AND mi.missing_scan_count = 0))
+ORDER BY ci.position ASC NULLS LAST, ci.added_at ASC;
+
+-- name: GetCollectionEpisodesOrdered :many
+-- The collection queue's downloaded episodes in collection order
+-- (plan item 2.3 "Play collection as sequence"): position-first, then
+-- added_at. INNER JOIN — the chain steps only through episodes the
+-- requester can see (RLS drops unsubscribed feeds) and that are
+-- downloaded (invariant 6: un-downloaded episodes are never tracked).
+SELECT mi.id, mi.title, mi.feed_position, mi.date_published
+FROM collection_items ci
+JOIN media_items mi ON ci.media_item_id = mi.id
+WHERE ci.collection_id = $1
+  AND mi.source = 'podcast_episode'
+  AND mi.download_state = 'downloaded'
+  AND mi.archived_at IS NULL
+  AND mi.missing_scan_count = 0
+ORDER BY ci.position ASC NULLS LAST, ci.added_at ASC;
+
 -- Get collections for book
 -- name: GetCollectionsForBook :many
 SELECT c.*
@@ -3936,8 +4026,8 @@ RETURNING *;
 DELETE FROM collection_items WHERE collection_id = $1;
 
 -- name: CreateSystemCollection :one
-INSERT INTO collections (user_id, name, description, icon, color, show_on_dashboard, query_type, priority, is_system_collection, auto_assign_rules)
+INSERT INTO collections (user_id, name, description, icon, color, show_on_dashboard, query_type, priority, is_system_collection)
-VALUES ($1, $2, $3, $4, $5, $6, $7, $8, true, 'null'::jsonb)
+VALUES ($1, $2, $3, $4, $5, $6, $7, $8, true)
 RETURNING *;
 
 -- name: ResetSystemCollectionMetadata :exec
@@ -3962,6 +4052,7 @@ INNER JOIN (
     ORDER BY media_item_id, last_read_at DESC
 ) rp ON rp.media_item_id = mi.id
 WHERE (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
 ORDER BY rp.last_read_at DESC
@@ -3970,6 +4061,7 @@ LIMIT sqlc.narg('limit');
 -- name: GetRecentlyAddedItems :many
 SELECT mi.* FROM media_items mi
 WHERE (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
 ORDER BY mi.imported_at DESC NULLS LAST, mi.created_at DESC
@@ -3985,6 +4077,7 @@ INNER JOIN (
     ORDER BY media_item_id, last_read_at DESC
 ) rp ON rp.media_item_id = mi.id
 WHERE (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
 ORDER BY rp.last_read_at DESC
@@ -3993,6 +4086,7 @@ LIMIT sqlc.narg('limit');
 -- name: GetNotStartedItems :many
 SELECT mi.* FROM media_items mi
 WHERE (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
   AND NOT EXISTS (
@@ -4049,19 +4143,6 @@ LIMIT sqlc.narg('limit');
 -- name: CountCollectionItems :one
 SELECT COUNT(*) FROM collection_items WHERE collection_id = $1;
 
--- Smart ("dynamic") collections: rules are evaluated LIVE at render, so
--- membership comes from the candidate book set instead of collection_items.
--- The set is scoped to the requesting user's visible libraries (family
--- lens enforced at the source) and live items only — the same filters the
--- static collection paths apply.
--- name: GetVisibleLiveMediaForRules :many
-SELECT mi.* FROM media_items mi
-WHERE (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
-  AND (sqlc.narg('visible_library_ids')::uuid[] IS NULL OR mi.library_id = ANY(sqlc.narg('visible_library_ids')::uuid[]))
-  AND mi.archived_at IS NULL
-  AND mi.missing_scan_count = 0
-ORDER BY mi.title ASC;
-
 -- Live item count for a STATIC collection (rule-free cards on the
 -- collections page); same live filter as GetCollectionItems, plus the
 -- future-hybrid excluded exclusion and an optional library scope.
@@ -4079,6 +4160,7 @@ SELECT mi.* FROM media_items mi
 WHERE (sqlc.narg('library_id')::uuid IS NULL OR mi.library_id = sqlc.narg('library_id')::uuid)
   AND mi.archived_at IS NULL
   AND mi.missing_scan_count = 0
+  AND (mi.library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9: collections picker)
 ORDER BY mi.created_at DESC;
 
 -- name: GetDistinctSeries :many
@@ -4111,6 +4193,7 @@ WHERE (sqlc.narg('library_id')::uuid IS NULL OR library_id = sqlc.narg('library_
   AND (sqlc.narg('visible_library_ids')::uuid[] IS NULL OR library_id = ANY(sqlc.narg('visible_library_ids')::uuid[]))
   AND archived_at IS NULL
   AND missing_scan_count = 0
+  AND (library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9: series pages)
   AND series IS NOT NULL AND series != ''
   -- Multi-field search (series page): the query matches the series
   -- NAME, any book's AUTHOR, or any book's TITLE in the group. It is
@@ -4171,6 +4254,7 @@ SELECT COUNT(*)::int FROM (
     AND (sqlc.narg('visible_library_ids')::uuid[] IS NULL OR library_id = ANY(sqlc.narg('visible_library_ids')::uuid[]))
     AND archived_at IS NULL
     AND missing_scan_count = 0
+    AND (library_type_name IS DISTINCT FROM 'podcasts') -- podcasts exclusion (plan §9: series pages)
     AND series IS NOT NULL AND series != ''
     AND (sqlc.narg('series_query')::text IS NULL OR
          series ILIKE '%' || sqlc.narg('series_query')::text || '%'
@@ -4590,3 +4674,481 @@ SET file_size = $2,
     updated_at = NOW()
 WHERE id = $1
 RETURNING id, library_id, title, author, isbn, description, file_path, file_size, mime_type, cover_image_path, series, series_number, tags, narrators, abridged, asin, date_published, publisher, contributors, language, edition, page_count, genre, copyright_year, goodreads_id, openlibrary_id, google_books_id, added_by_admin_id, created_at, imported_at, updated_at, format_group, format_mimetype, is_reflowable, has_fixed_layout, total_characters, chapter_count, entitlement_id, revision_number, kobo_content_id, kobo_metadata, manga_type, reading_direction, series_count, volume, imprint, age_rating, web_url, story_arc, is_black_and_white, metadata_notes, community_rating, alternate_info, scan_information, summary, metadata_overrides, missing_scan_count, archived_at, chapter_metadata, library_type_name, audio_metadata, tags_search, contributors_search, narrators_search, file_sha256, opf_identifier, opf_uuid, hash_confidence;
+
+-- ============================================================================
+-- PODCASTS / RSS FEEDS (plan ~/.opencode/plan/podcast-integration.md §6/§8)
+-- Visibility is RLS-enforced (schema policy block) — NO query here carries
+-- visibility clauses. Feed reads run under the acting user's identity
+-- (subscriber-or-admin policy); the fetcher/worker run as first-admin.
+-- The fetcher/download worker is the SOLE writer of feed/episode metadata;
+-- users write only podcast_subscriptions rows.
+-- ============================================================================
+
+-- name: CreatePodcastFeed :one
+-- Add-feed upsert keyed on the per-library UNIQUE(library_id, feed_url):
+-- a second subscriber adding the same URL attaches to the existing row
+-- instead of erroring. DO UPDATE (not DO NOTHING) so the RETURNING works
+-- even when the conflicting row is invisible to the actor under RLS
+-- (non-admin creators cannot SELECT other users' feeds). Metadata columns
+-- are filled by the fetcher right after; this only establishes identity.
+INSERT INTO podcast_feeds (library_id, feed_url, title, created_by_admin_id)
+VALUES ($1, $2, $3, $4)
+ON CONFLICT (library_id, feed_url) DO UPDATE SET updated_at = NOW()
+RETURNING *;
+
+-- name: GetPodcastFeed :one
+SELECT * FROM podcast_feeds WHERE id = $1;
+
+-- name: ListMyPodcastFeeds :many
+-- The podcasts page: the acting user's subscriptions with feed metadata
+-- and the user-owned subscription fields. Facet values derive in Go.
+SELECT f.*,
+       s.display_name,
+       s.auto_download,
+       s.auto_download_count,
+       s.created_at AS subscribed_at
+FROM podcast_subscriptions s
+JOIN podcast_feeds f ON f.id = s.feed_id
+WHERE s.user_id = $1
+ORDER BY COALESCE(s.display_name, f.title) ASC;
+
+-- name: GetPodcastSubscription :one
+SELECT * FROM podcast_subscriptions WHERE user_id = $1 AND feed_id = $2;
+
+-- name: SubscribePodcastFeed :one
+INSERT INTO podcast_subscriptions (user_id, feed_id)
+VALUES ($1, $2)
+ON CONFLICT (user_id, feed_id) DO NOTHING
+RETURNING *;
+
+-- name: UpdatePodcastSubscription :one
+-- User-owned fields only (D9): display name + auto-download policy.
+UPDATE podcast_subscriptions
+SET display_name = $3,
+    auto_download = $4,
+    auto_download_count = $5
+WHERE user_id = $1 AND feed_id = $2
+RETURNING *;
+
+-- name: DeletePodcastSubscription :execrows
+-- Unsubscribe NEVER deletes feed/episode/file state (D6).
+DELETE FROM podcast_subscriptions WHERE user_id = $1 AND feed_id = $2;
+
+-- name: ListFeedSubscribers :many
+-- Download-worker policy union (D6): most aggressive policy wins for
+-- what arrives. Runs under an admin identity (background job).
+SELECT user_id, auto_download, auto_download_count
+FROM podcast_subscriptions
+WHERE feed_id = $1;
+
+-- name: CountPodcastFeedSubscribers :one
+SELECT COUNT(*) FROM podcast_subscriptions WHERE feed_id = $1;
+
+-- name: ListAllPodcastFeedsAdmin :many
+-- Admin all-podcasts view (§0 item 8 / Phase-4 item 1): EVERY feed from
+-- EVERY user with its maintenance aggregates. Runs under the first-
+-- admin identity (invariant 12 — same contract as the poller). Sorted
+-- by disk usage DESC: the maintenance question is "what's eating the
+-- disk", so the hogs surface first.
+SELECT f.id, f.title, f.host, f.last_fetched_at, f.library_id,
+       l.name AS library_name,
+       (SELECT COUNT(*) FROM media_items m
+         WHERE m.source_feed_id = f.id AND m.source = 'podcast_episode') AS episode_total,
+       (SELECT COUNT(*) FROM media_items m
+         WHERE m.source_feed_id = f.id AND m.source = 'podcast_episode'
+           AND m.download_state = 'downloaded') AS episode_downloaded,
+       (SELECT COALESCE(SUM(m.file_size), 0)::BIGINT FROM media_items m
+         WHERE m.source_feed_id = f.id AND m.source = 'podcast_episode'
+           AND m.download_state = 'downloaded') AS disk_bytes,
+       (SELECT COUNT(*) FROM podcast_subscriptions s
+         WHERE s.feed_id = f.id) AS subscriber_count,
+       (SELECT COALESCE(string_agg(u.username, ', ' ORDER BY u.username), '')::TEXT
+          FROM podcast_subscriptions s
+          JOIN users u ON u.id = s.user_id
+         WHERE s.feed_id = f.id) AS subscriber_names
+FROM podcast_feeds f
+JOIN libraries l ON l.id = f.library_id
+ORDER BY 9 DESC, f.title ASC;
+
+-- name: ResetPodcastEpisodeFileDeleted :exec
+-- Admin file deletion (§0 item 8): ONE downloaded episode's disk file
+-- was removed — reset the row to pending and clear the download-derived
+-- facts. file_path is deliberately KEPT: it is the fetch-maintained
+-- layout (written at fetch time, not download time), so a re-download
+-- lands on the same path. The poll upsert never touches download_state,
+-- so a deleted file stays deleted (invariant 2: polling never deletes;
+-- admins delete knowingly).
+UPDATE media_items
+SET download_state = 'pending',
+    download_error = NULL,
+    download_attempts = 0,
+    file_size = NULL,
+    mime_type = NULL,
+    cover_image_path = NULL,
+    audio_metadata = NULL,
+    chapter_count = NULL,
+    updated_at = NOW()
+WHERE id = $1 AND source = 'podcast_episode' AND download_state = 'downloaded';
+
+-- name: ResetPodcastFeedFilesDeleted :execrows
+-- Same reset for EVERY downloaded episode of a feed (per-feed
+-- "Delete files" action on the admin view).
+UPDATE media_items
+SET download_state = 'pending',
+    download_error = NULL,
+    download_attempts = 0,
+    file_size = NULL,
+    mime_type = NULL,
+    cover_image_path = NULL,
+    audio_metadata = NULL,
+    chapter_count = NULL,
+    updated_at = NOW()
+WHERE source_feed_id = $1 AND source = 'podcast_episode'
+  AND download_state = 'downloaded';
+
+-- name: ListPodcastFeedsForPolling :many
+-- Poller worklist (admin identity): every feed with its validators and
+-- the back-off flag (is_complete feeds poll rarely, never disappear).
+SELECT * FROM podcast_feeds ORDER BY last_fetched_at ASC NULLS FIRST;
+
+-- name: UpdatePodcastFeedFetch :one
+-- Post-fetch bookkeeping: validators for conditional GETs, the fetch
+-- timestamp, and the feed's own last-build marker.
+UPDATE podcast_feeds
+SET etag = $2,
+    last_modified = $3,
+    last_fetched_at = NOW(),
+    last_build_date = $4,
+    updated_at = NOW()
+WHERE id = $1
+RETURNING *;
+
+-- name: UpdatePodcastFeedMetadata :one
+-- Fetcher-only full metadata refresh (parse result). podcast_guid is
+-- set once a feed declares it; new_feed_url handling is a dedicated
+-- query below.
+UPDATE podcast_feeds
+SET title = $2,
+    description = $3,
+    host = $4,
+    publisher = $5,
+    owner_name = $6,
+    owner_email = $7,
+    language = $8,
+    tags = $9,
+    explicit = $10,
+    feed_type = $11,
+    medium = $12,
+    is_complete = $13,
+    podcast_guid = COALESCE($14, podcast_guid),
+    updated_at = NOW()
+WHERE id = $1
+RETURNING *;
+
+-- name: UpdatePodcastFeedArtwork :exec
+UPDATE podcast_feeds SET artwork_path = $2, updated_at = NOW() WHERE id = $1;
+
+-- name: UpdatePodcastFeedRedirect :exec
+-- itunes:new-feed-url (or a permanent HTTP redirect chain): the feed has
+-- MOVED. Rewrite feed_url so future polls hit the new location; the row
+-- (and every subscription/episode) stays put.
+UPDATE podcast_feeds
+SET feed_url = $2,
+    updated_at = NOW()
+WHERE id = $1;
+
+-- name: GetPodcastFeedByGuid :one
+-- URL-migration dedupe: same canonical guid within the library reuses
+-- the existing row.
+SELECT * FROM podcast_feeds
+WHERE library_id = $1 AND podcast_guid = $2
+LIMIT 1;
+
+-- name: DeletePodcastFeed :execrows
+-- Admin-only removal (handler checks role); episodes cascade via
+-- media_items.source_feed_id ON DELETE CASCADE.
+DELETE FROM podcast_feeds WHERE id = $1;
+
+-- name: CountPodcastLibrariesForUser :one
+-- Sidebar/page gate (D13): >=1 podcasts-type library the user can see
+-- (COALESCE(is_visible,true) semantics — mirrors GetUserVisibleLibraries).
+SELECT COUNT(*)
+FROM libraries l
+JOIN library_types lt ON lt.id = l.library_type_id AND lt.name = 'podcasts'
+LEFT JOIN library_visibility lv ON lv.library_id = l.id AND lv.user_id = $1
+WHERE COALESCE(lv.is_visible, true) = true;
+
+-- name: GetPodcastsLibraryForUser :many
+-- The podcasts library/ies visible to the user (add-feed target list;
+-- a self-hosted instance typically has exactly one).
+SELECT l.id, l.name
+FROM libraries l
+JOIN library_types lt ON lt.id = l.library_type_id AND lt.name = 'podcasts'
+LEFT JOIN library_visibility lv ON lv.library_id = l.id AND lv.user_id = $1
+WHERE COALESCE(lv.is_visible, true) = true
+ORDER BY l.created_at ASC;
+
+-- name: GetPodcastLibraryFolder :many
+-- The download root (purpose='podcast') of a podcasts library.
+SELECT lf.*
+FROM library_folders lf
+WHERE lf.library_id = $1 AND lf.purpose = 'podcast';
+
+-- name: CreatePodcastLibraryFolder :one
+INSERT INTO library_folders (library_id, folder_path, purpose)
+VALUES ($1, $2, 'podcast')
+RETURNING *;
+
+-- ============================================================================
+-- Podcast episodes — media_items rows with source='podcast_episode'
+-- ============================================================================
+
+-- name: UpsertPodcastEpisode :one
+-- Fetcher-only upsert keyed on the partial unique index
+-- (source_feed_id, episode_guid). DO UPDATE refreshes FETCH-OWNED
+-- metadata only. Write-after-success guard (AGENTS.md gotcha #7):
+-- download state/error/attempts are NEVER touched here, and the
+-- downloaded file's facts (path/size/mime/cover) survive renumbering —
+-- only un-downloaded rows follow the re-derived file_path.
+-- authors_search mirrors authors (trigram search column).
+INSERT INTO media_items (
+    library_id, title, subtitle, description, authors, authors_search,
+    file_path, file_size, mime_type, cover_image_path, explicit,
+    date_published, language, tags, genre,
+    format_group, format_mimetype, is_reflowable, library_type_name,
+    source, source_feed_id, episode_guid, feed_position,
+    season_number, episode_number, enclosure_url,
+    declared_duration_seconds, transcript_url, chapters_url,
+    hosts, guests
+) VALUES (
+    $1, $2, $3, $4, $5, $6,
+    $7, $8, $9, $10, $11,
+    $12, $13, $14, $15,
+    'audiobook', $9, FALSE, 'podcasts',
+    'podcast_episode', $16, $17, $18,
+    $19, $20, $21,
+    $22, $23, $24,
+    $25, $26
+)
+ON CONFLICT (source_feed_id, episode_guid) WHERE source_feed_id IS NOT NULL DO UPDATE SET
+    title = EXCLUDED.title,
+    subtitle = EXCLUDED.subtitle,
+    description = EXCLUDED.description,
+    authors = EXCLUDED.authors,
+    authors_search = EXCLUDED.authors_search,
+    explicit = EXCLUDED.explicit,
+    date_published = EXCLUDED.date_published,
+    language = EXCLUDED.language,
+    tags = EXCLUDED.tags,
+    genre = EXCLUDED.genre,
+    season_number = EXCLUDED.season_number,
+    episode_number = EXCLUDED.episode_number,
+    enclosure_url = EXCLUDED.enclosure_url,
+    declared_duration_seconds = EXCLUDED.declared_duration_seconds,
+    transcript_url = EXCLUDED.transcript_url,
+    chapters_url = EXCLUDED.chapters_url,
+    hosts = EXCLUDED.hosts,
+    guests = EXCLUDED.guests,
+    feed_position = EXCLUDED.feed_position,
+    cover_image_path = CASE WHEN media_items.download_state = 'downloaded' THEN media_items.cover_image_path ELSE EXCLUDED.cover_image_path END,
+    file_path = CASE WHEN media_items.download_state = 'downloaded' THEN media_items.file_path ELSE EXCLUDED.file_path END,
+    file_size = CASE WHEN media_items.download_state = 'downloaded' THEN media_items.file_size ELSE EXCLUDED.file_size END,
+    mime_type = CASE WHEN media_items.download_state = 'downloaded' THEN media_items.mime_type ELSE EXCLUDED.mime_type END,
+    updated_at = NOW()
+RETURNING *;
+
+-- name: ListPodcastEpisodes :many
+-- Feed detail / sequence source. RLS scopes to subscriber-or-admin;
+-- ordering: native position, then pub date, then guid (stable across
+-- re-fetches for historical rows whose position went stale).
+SELECT * FROM media_items
+WHERE source_feed_id = $1 AND source = 'podcast_episode'
+ORDER BY feed_position ASC NULLS LAST, date_published ASC NULLS LAST, episode_guid ASC;
+
+-- name: GetPodcastEpisode :one
+SELECT * FROM media_items
+WHERE id = $1 AND source = 'podcast_episode';
+
+-- name: SetPodcastEpisodeDownloadQueued :execrows
+-- Enqueue: pending|failed -> queued, stamped with the NEXT QUEUE TICKET
+-- (§0 items 4/5) so claim order = enqueue order (the effective
+-- listening order the enqueue loop applied). Attempts increment at
+-- enqueue so a crashed worker still counts the try.
+UPDATE media_items
+SET download_state = 'queued',
+    download_queue_pos = nextval('podcast_queue_seq'),
+    download_error = NULL,
+    download_attempts = download_attempts + 1,
+    updated_at = NOW()
+WHERE id = $1
+  AND source = 'podcast_episode'
+  AND download_state IN ('pending', 'failed');
+
+-- SetPodcastEpisodeDownloading RETIRED (§0 item 4): the pair
+-- GetNextQueuedEpisode + SetPodcastEpisodeDownloading raced under
+-- concurrency; ClaimQueuedPodcastEpisode does both atomically.
+
+-- name: MarkPodcastEpisodeDownloaded :one
+-- SUCCESS-only write (invariant 7): the file is on disk and probed.
+-- ffprobe facts (real duration via audio_metadata, precise size/mime)
+-- reconcile the feed's declared values.
+UPDATE media_items
+SET download_state = 'downloaded',
+    download_error = NULL,
+    file_path = $2,
+    file_size = $3,
+    mime_type = $4,
+    cover_image_path = COALESCE($5, cover_image_path),
+    audio_metadata = $6,
+    chapter_count = $7,
+    updated_at = NOW()
+WHERE id = $1 AND source = 'podcast_episode'
+RETURNING *;
+
+-- name: MarkPodcastEpisodeFailed :exec
+-- Failure/rejection write: state + reason + whether a human already
+-- intervened (manual_retry resets the auto-retry budget). Existing
+-- metadata/file facts are never cleared.
+UPDATE media_items
+SET download_state = 'failed',
+    download_error = $2,
+    download_attempts = CASE WHEN $3::boolean THEN 0 ELSE download_attempts END,
+    updated_at = NOW()
+WHERE id = $1 AND source = 'podcast_episode';
+
+-- name: SetPodcastEpisodePending :exec
+-- Manual retry (same endpoint as download): reset to pending so the
+-- worker re-claims it; the reason text clears with the fresh attempt.
+UPDATE media_items
+SET download_state = 'pending',
+    download_error = NULL,
+    download_attempts = 0,
+    updated_at = NOW()
+WHERE id = $1 AND source = 'podcast_episode';
+
+-- name: ListUndownloadedEpisodes :many
+-- Download All + policy union worklist.
+SELECT * FROM media_items
+WHERE source_feed_id = $1
+  AND source = 'podcast_episode'
+  AND download_state <> 'downloaded'
+ORDER BY feed_position ASC NULLS LAST, date_published ASC NULLS LAST, episode_guid ASC;
+
+-- name: CountUndownloadedEpisodes :one
+SELECT COUNT(*) FROM media_items
+WHERE source_feed_id = $1
+  AND source = 'podcast_episode'
+  AND download_state <> 'downloaded';
+
+-- name: CountDownloadedEpisodes :one
+SELECT COUNT(*) FROM media_items
+WHERE source_feed_id = $1
+  AND source = 'podcast_episode'
+  AND download_state = 'downloaded';
+
+-- name: SearchMyPodcastEpisodes :many
+-- Subscription-scoped episode search (D14): title + description cover
+-- un-downloaded AND historical rows (fetcher persists metadata-only).
+-- RLS already restricts to the caller's subscriptions — the explicit
+-- source filter keeps scan items out of podcast search.
+SELECT * FROM media_items
+WHERE source = 'podcast_episode'
+  AND (sqlc.arg('search_query')::text = '' OR title ILIKE '%' || sqlc.arg('search_query')::text || '%' OR description ILIKE '%' || sqlc.arg('search_query')::text || '%')
+ORDER BY feed_position ASC NULLS LAST, date_published ASC NULLS LAST
+LIMIT 50;
+
+-- name: SearchMyPodcastFeeds :many
+-- Feed-title AND subscription-display-name search (D14) + the facet
+-- fields the podcasts page filters on.
+SELECT f.*,
+       s.display_name,
+       s.auto_download,
+       s.auto_download_count,
+       s.created_at AS subscribed_at
+FROM podcast_subscriptions s
+JOIN podcast_feeds f ON f.id = s.feed_id
+WHERE s.user_id = sqlc.arg('user_id')
+  AND (sqlc.arg('search_query')::text = '' OR f.title ILIKE '%' || sqlc.arg('search_query')::text || '%'
+       OR COALESCE(s.display_name, '') ILIKE '%' || sqlc.arg('search_query')::text || '%'
+       OR COALESCE(f.host, '') ILIKE '%' || sqlc.arg('search_query')::text || '%'
+       OR COALESCE(f.publisher, '') ILIKE '%' || sqlc.arg('search_query')::text || '%')
+ORDER BY COALESCE(s.display_name, f.title) ASC;
+
+-- name: ListQueuedPodcastEpisodes :many
+-- Fair-share candidate scan (§0 item 4): the download pool claims in
+-- TICKET order (download_queue_pos = SEQUENCE nextval at enqueue — the
+-- enqueue loop's ordering IS the download order, item 5). The worker
+-- groups these per feed and prefers the least-recently-served feed.
+SELECT id, source_feed_id FROM media_items
+WHERE source = 'podcast_episode' AND download_state = 'queued'
+ORDER BY download_queue_pos ASC NULLS LAST;
+
+-- name: ClaimQueuedPodcastEpisode :one
+-- Atomic claim (§0 item 4): converts queued -> downloading in ONE
+-- statement. Two claimants racing on the same row: exactly one UPDATE
+-- matches download_state='queued'; the loser gets no rows (ErrNoRows).
+-- This is the SKIP-LOCKED guarantee without a second round-trip.
+UPDATE media_items
+SET download_state = 'downloading', updated_at = NOW()
+WHERE id = $1
+  AND source = 'podcast_episode'
+  AND download_state = 'queued'
+RETURNING media_items.*;
+
+-- name: CountQueuedEpisodes :one
+SELECT COUNT(*) FROM media_items
+WHERE source = 'podcast_episode' AND download_state IN ('queued', 'downloading');
+
+-- name: CountPodcastEpisodesForFeed :one
+SELECT COUNT(*) FROM media_items
+WHERE source_feed_id = $1 AND source = 'podcast_episode';
+
+-- ============================================================================
+-- GENERIC reader_settings KV (podcast order override, plan D5: key
+-- podcast_order:<feed_id>). The pre-existing reader_settings queries are
+-- hard-keyed to 'reader_settings'; these two are the generic form.
+-- ============================================================================
+
+-- name: GetUserSettingByKey :one
+SELECT setting_value FROM reader_settings
+WHERE user_id = $1 AND setting_key = $2;
+
+-- name: UpsertUserSettingByKey :exec
+INSERT INTO reader_settings (user_id, setting_key, setting_value)
+VALUES ($1, $2, $3)
+ON CONFLICT (user_id, setting_key)
+DO UPDATE SET setting_value = EXCLUDED.setting_value, updated_at = NOW();
+
+-- name: DeleteUserSettingByKey :exec
+DELETE FROM reader_settings WHERE user_id = $1 AND setting_key = $2;
+
+-- name: GetPodcastEpisodeProgress :many
+-- Per-user progress for a feed's episodes (row badges + the Phase-2
+-- continue fields). reading_progress is UNIQUE per (media_item, user).
+SELECT rp.media_item_id,
+       rp.percentage,
+       rp.last_read_at
+FROM reading_progress rp
+JOIN media_items mi ON mi.id = rp.media_item_id
+WHERE rp.user_id = $1
+  AND mi.source_feed_id = $2
+  AND mi.source = 'podcast_episode';
+
+-- name: GetPodcastContinueForUser :many
+-- Continue aggregation (plan 2.5): the caller's IN-PROGRESS episodes
+-- (0 < pct < finishedThreshold) across ALL subscribed feeds, most
+-- recent first. Go groups per feed — the first row per feed IS the
+-- continue target.
+SELECT mi.source_feed_id,
+       mi.id,
+       mi.title,
+       mi.feed_position,
+       rp.percentage,
+       rp.last_read_at
+FROM reading_progress rp
+JOIN media_items mi ON mi.id = rp.media_item_id
+WHERE rp.user_id = $1
+  AND mi.source = 'podcast_episode'
+  AND rp.percentage > 0
+  AND rp.percentage < 0.995
+ORDER BY rp.last_read_at DESC;

internal/database/settings_registry.go

@@ -25,6 +25,7 @@ import (
 // SettingType enumerates the value types stored in system_settings.setting_type.
 const (
 	SettingTypeInt        = "int"
+	SettingTypeFloat      = "float"
 	SettingTypeBool       = "bool"
 	SettingTypeString     = "string"
 	SettingTypeStringList = "string_list"
@@ -82,6 +83,9 @@ var SettingDefaults = []SettingDefault{
 	{Key: "sync_queue_batch_size", Value: "50", Type: SettingTypeInt, Min: "1", Max: "10000", RequiresRestart: true, Category: "sync", Group: "Sync Queue", Description: "Maximum items processed per sync queue flush"},
 	{Key: "worker_pool_size", Value: "3", Type: SettingTypeInt, Min: "1", Max: "100", RequiresRestart: true, Category: "performance", Group: "Worker Pool", Description: "Number of background worker goroutines"},
 	{Key: "worker_queue_cap", Value: "100", Type: SettingTypeInt, Min: "1", Max: "10000", RequiresRestart: true, Category: "performance", Group: "Worker Pool", Description: "Background worker job queue capacity"},
+	{Key: "podcast_poll_interval_seconds", Value: "3600", Type: SettingTypeInt, Min: "300", Max: "604800", Category: "podcasts", Group: "Podcasts", Description: "How often to poll podcast feeds for new episodes (seconds; complete feeds back off)"},
+	{Key: "podcast_download_concurrency", Value: "4", Type: SettingTypeInt, Min: "1", Max: "16", Category: "podcasts", Group: "Podcasts", Description: "Podcast episode downloads in flight server-wide (the connection is the limit; fair-share spreads them across feeds)"},
+	{Key: "podcast_download_speed_mbps", Value: "0", Type: SettingTypeFloat, Min: "0", Max: "10000", Category: "podcasts", Group: "Podcasts", Description: "Shared bandwidth cap for podcast downloads in megabits/second (0 = unlimited)"},
 }
 
 // defaultBy indexes SettingDefaults by key for O(1) lookup.
@@ -172,6 +176,26 @@ func (r *SettingsRegistry) getInt(key string) int {
 	return n
 }
 
+func (r *SettingsRegistry) getFloat(key string) float64 {
+	d := defaultBy[key]
+	v := r.raw(key)
+	f, err := strconv.ParseFloat(v, 64)
+	if err != nil {
+		f, _ = strconv.ParseFloat(d.Value, 64)
+	}
+	if d.Min != "" {
+		if mn, err := strconv.ParseFloat(d.Min, 64); err == nil && f < mn {
+			f = mn
+		}
+	}
+	if d.Max != "" {
+		if mx, err := strconv.ParseFloat(d.Max, 64); err == nil && f > mx {
+			f = mx
+		}
+	}
+	return f
+}
+
 func (r *SettingsRegistry) getBool(key string) bool {
 	v := r.raw(key)
 	b, err := strconv.ParseBool(v)
@@ -188,6 +212,24 @@ func (r *SettingsRegistry) ScanPollInterval() time.Duration {
 	return time.Duration(r.getInt("scan_poll_interval_seconds")) * time.Second
 }
 
+// PodcastPollInterval is how often the poller checks feeds, as a
+// duration (live setting; the poller re-reads it every cycle — writes
+// go through the settings API path, AGENTS.md gotcha #2).
+func (r *SettingsRegistry) PodcastPollInterval() time.Duration {
+	return time.Duration(r.getInt("podcast_poll_interval_seconds")) * time.Second
+}
+
+// PodcastDownloadConcurrency: episode downloads in flight server-wide
+// (the connection is the limit; fair-share spreads them across feeds).
+func (r *SettingsRegistry) PodcastDownloadConcurrency() int {
+	return r.getInt("podcast_download_concurrency")
+}
+
+// PodcastDownloadSpeedMbps: shared bandwidth cap (0 = unlimited).
+func (r *SettingsRegistry) PodcastDownloadSpeedMbps() float64 {
+	return r.getFloat("podcast_download_speed_mbps")
+}
+
 // AutoScanEnabled reports whether auto-scanning is on.
 func (r *SettingsRegistry) AutoScanEnabled() bool { return r.getBool("auto_scan_enabled") }
 

internal/database/settings_registry_test.go

@@ -36,6 +36,21 @@ func TestSettingDefaults(t *testing.T) {
 			if _, err := strconv.ParseBool(d.Value); err != nil {
 				t.Errorf("bool setting %s default %q is not a bool", d.Key, d.Value)
 			}
+		case SettingTypeFloat:
+			f, err := strconv.ParseFloat(d.Value, 64)
+			if err != nil {
+				t.Errorf("float setting %s default %q is not a float", d.Key, d.Value)
+			}
+			if d.Min != "" {
+				if mn, err := strconv.ParseFloat(d.Min, 64); err == nil && f < mn {
+					t.Errorf("float setting %s default %q below min %q", d.Key, d.Value, d.Min)
+				}
+			}
+			if d.Max != "" {
+				if mx, err := strconv.ParseFloat(d.Max, 64); err == nil && f > mx {
+					t.Errorf("float setting %s default %q above max %q", d.Key, d.Value, d.Max)
+				}
+			}
 		case SettingTypeString:
 			if d.Value == "" {
 				t.Errorf("string setting %s has empty default", d.Key)

internal/docs/handler.go

@@ -5,6 +5,7 @@ import (
 	"fmt"
 	"io/fs"
 	"os"
+	"path/filepath"
 	"strings"
 
 	"bookhoard/templates"
@@ -24,7 +25,23 @@ type DocsHandler struct {
 	docsFS   fs.FS
 }
 
+// resolveDocsPath — the docs folder lives at the repo root, but `go test`
+// runs test binaries from the PACKAGE directory (/cmd/server/tests), so a
+// plain relative "docs" resolves to a folder that doesn't exist there and
+// every /docs route 404s in tests (2026-11-01). Walk upward, bounded,
+// until the folder is real; prod (/app) matches on the first try.
+func resolveDocsPath(p string) string {
+	for i := 0; i < 6; i++ {
+		if fi, err := os.Stat(filepath.Join(p, "index.md")); err == nil && !fi.IsDir() {
+			return p
+		}
+		p = filepath.Join("..", p)
+	}
+	return "docs"
+}
+
 func NewDocsHandler(docsPath string) *DocsHandler {
+	docsPath = resolveDocsPath(docsPath)
 	// Create goldmark renderer with syntax highlighting
 	md := goldmark.New(
 		goldmark.WithExtensions(
@@ -74,11 +91,24 @@ func (h *DocsHandler) LoadDocument(docPath string) (*templates.Document, error)
 		}
 	}
 
-	// Read the markdown file
+	// Read the markdown file. Folder URLs resolve to their index page:
+	// /docs/user/filters → user/filters/index.md (2026-11-01 — the
+	// filter manual is a folder: guide + cookbook + generated field
+	// reference).
 	content, err := fs.ReadFile(h.docsFS, docPath)
 	if err != nil {
+		if !strings.HasSuffix(docPath, "index.md") {
+			alt := strings.TrimSuffix(docPath, ".md") + "/index.md"
+			if altContent, altErr := fs.ReadFile(h.docsFS, alt); altErr == nil {
+				docPath = alt
+				content = altContent
+			} else {
 				return nil, fmt.Errorf("failed to read document: %w", err)
 			}
+		} else {
+			return nil, fmt.Errorf("failed to read document: %w", err)
+		}
+	}
 
 	// Convert markdown to HTML
 	var buf bytes.Buffer

internal/docs/navigation.go

@@ -321,10 +321,23 @@ func (h *DocsHandler) BuildNavigation(currentPath string) *templates.Navigation
 	return nav
 }
 
-// containsURL reports whether currentPath is one of the given doc URLs.
+// normalizeDocURL — folder-index pages SERVE at their directory URL
+// (/docs/user/filters) while their nav entry carries the /index suffix
+// (/docs/user/filters/index): same page, two spellings. Both sidebar
+// checks (expansion + highlight) compare normalized forms, so folder
+// pages expand and highlight like file pages — no per-page special
+// cases (2026-11-01, the filter manual's folder).
+func normalizeDocURL(u string) string {
+	u = strings.TrimRight(u, "/")
+	return strings.TrimSuffix(u, "/index")
+}
+
+// containsURL reports whether currentPath is one of the given doc URLs
+// (folder-index URLs normalized — see normalizeDocURL).
 func containsURL(currentPath string, urls []string) bool {
+	current := normalizeDocURL(currentPath)
 	for _, u := range urls {
-		if currentPath == u {
+		if current == normalizeDocURL(u) {
 			return true
 		}
 	}

internal/docs/navigation_test.go

@@ -4,6 +4,8 @@ import (
 	"os"
 	"path/filepath"
 	"testing"
+
+	"bookhoard/templates"
 )
 
 func TestPlaceDoc(t *testing.T) {
@@ -205,3 +207,41 @@ func TestBuildNavigationNothingCurrent(t *testing.T) {
 		t.Errorf("expected zero sections open when nothing matches (docs home / 404 renders fully closed), got %d", open)
 	}
 }
+
+// TestBuildNavigationFolderIndex (2026-11-01): folder-index pages serve
+// at their directory URL (/docs/user/filters) while their nav entries
+// carry /index (/docs/user/filters/index) — the sidebar must EXPAND the
+// containing section/group and HIGHLIGHT the index item for the folder
+// URL, exactly like a file page. The filter manual's folder is the
+// first page of this shape.
+func TestBuildNavigationFolderIndex(t *testing.T) {
+	// The REAL docs tree: the Filters group only exists because the
+	// filter manual (guide/cookbook/fields) lives in user/filters/.
+	handler := NewDocsHandler("../../docs")
+	nav := handler.BuildNavigation("/docs/user/filters")
+
+	var userGuide *templates.NavSection
+	for i := range nav.Sections {
+		if nav.Sections[i].Title == "User Guide" {
+			userGuide = &nav.Sections[i]
+		}
+	}
+	if userGuide == nil {
+		t.Fatal("User Guide section missing")
+	}
+	if userGuide.Collapsed {
+		t.Error("User Guide section must expand for /docs/user/filters")
+	}
+	var filters *templates.NavGroup
+	for i := range userGuide.Groups {
+		if userGuide.Groups[i].Title == "Filters" {
+			filters = &userGuide.Groups[i]
+		}
+	}
+	if filters == nil {
+		t.Fatal("Filters group missing from User Guide")
+	}
+	if filters.Collapsed {
+		t.Error("Filters group must expand for its folder URL")
+	}
+}

internal/docs/schema.go

@@ -49,7 +49,7 @@ var Schemas = map[string]Schema{
 		Response: json.RawMessage(`{
 			"id": "", "user_id": "", "name": "", "description": "",
 			"color": "", "icon": "", "collection_type": "",
-			"auto_assign_rules": null, "view_settings": {},
+			"view_settings": {},
 			"book_count": 0, "created_at": ""
 		}`),
 	},
@@ -62,7 +62,7 @@ var Schemas = map[string]Schema{
 	"GET /api/collections/:id": {
 		Response: json.RawMessage(`{
 			"id": "", "user_id": "", "name": "", "description": "",
-			"color": "", "icon": "", "auto_assign_rules": null,
+			"color": "", "icon": "",
 			"view_settings": {}, "books": [], "book_count": 0,
 			"is_smart": false, "up_next": [], "created_at": ""
 		}`),
@@ -71,7 +71,7 @@ var Schemas = map[string]Schema{
 		Request: handlers.UpdateCollectionRequest{},
 		Response: json.RawMessage(`{
 			"id": "", "user_id": "", "name": "", "description": "",
-			"color": "", "icon": "", "auto_assign_rules": null,
+			"color": "", "icon": "",
 			"view_settings": {}, "created_at": ""
 		}`),
 	},
@@ -601,15 +601,14 @@ var Schemas = map[string]Schema{
 		Request:  json.RawMessage(`{"item_ids": []}`),
 		Response: json.RawMessage(`{"status": "reordered"}`),
 	},
-	"PUT /api/collections/:id/rules/:ruleId": {
+	"PUT /api/collections/:id/view-settings": {
-		// Handler is SetRuleEnabled: toggle only, request {enabled required},
+		// Merges chain_order into the collection's view_settings JSONB
-		// response {collection_id, is_smart}.
+		// (playlist toggles share the doc and are preserved).
-		Request: json.RawMessage(`{"enabled": false}`),
+		Request: json.RawMessage(`{"chain_order": "newest"}`),
 		Response: json.RawMessage(`{
-			"collection_id": "", "is_smart": false
+			"id": "", "chain_order": "", "view_settings": {}
 		}`),
 	},
-
 	// --- Beam batch 7: scanner ---
 	"POST /api/scanner/scan": {
 		Request: handlers.ScanLibraryRequest{},
@@ -835,22 +834,6 @@ var Schemas = map[string]Schema{
 	"DELETE /api/devices/:id/file-aliases/:aliasId": {
 		Response: json.RawMessage(`{"message": ""}`),
 	},
-	"POST /api/collections/:id/rules": {
-		// First rule converts a static collection to smart.
-		Request: handlers.AddRuleRequest{},
-		Response: json.RawMessage(`{
-			"collection_id": "", "rule": {}, "is_smart": false
-		}`),
-	},
-	"DELETE /api/collections/:id/rules/:ruleId": {
-		Response: json.RawMessage(`{
-			"collection_id": "", "is_smart": false
-		}`),
-	},
-	"POST /api/collections/test-rules": {
-		Request:  handlers.TestRulesRequest{},
-		Response: json.RawMessage(`{"matches": [], "total": 0}`),
-	},
 	"POST /api/collections/:id/books": {
 		// Bulk add: {operations: [{collection_id, book_ids:[...]}]}.
 		Request: json.RawMessage(`{

internal/handlers/auth.go

@@ -6,6 +6,7 @@ package handlers
 import (
 	"bookhoard/internal/database"
 	"bookhoard/internal/middleware"
+	"bookhoard/internal/rls"
 	"bookhoard/internal/setupstatus"
 	"context"
 	"errors"
@@ -15,6 +16,7 @@ import (
 	"math"
 	"net/http"
 	"os"
+	"strconv"
 	"strings"
 	"time"
 
@@ -125,6 +127,10 @@ type UserProfile struct {
 	DefaultAudioSpeed float64 `json:"default_audio_speed,omitempty"`
 	// Whether the web reader/player should hold a Screen Wake Lock.
 	KeepScreenAwake bool `json:"keep_screen_awake"`
+	// Resume rewind: seconds before the saved position the audio
+	// player starts (open-resume and every genuine unpause). 0 is a
+	// real value (exact resume) — never omitted.
+	AudioResumeRewind int `json:"audio_resume_rewind_seconds"`
 }
 
 type ProfileUpdateRequest struct {
@@ -141,6 +147,11 @@ type ProfileUpdateRequest struct {
 	// Per the SPEEDS list bounds; the profile select always sends it,
 	// JSON clients omit it (0 = untouched).
 	DefaultAudioSpeed float64 `json:"default_audio_speed,omitempty" form:"default_audio_speed" validate:"omitempty,min=0.25,max=4"`
+	// Resume rewind seconds, as the form select sends it ("0"/"5"/"10").
+	// The string shape (keep_screen_awake precedent) lets the form
+	// select an explicit 0=Off while JSON clients that omit it decode
+	// as "" and leave the column untouched.
+	AudioResumeRewind string `json:"audio_resume_rewind_seconds,omitempty" form:"audio_resume_rewind_seconds" validate:"omitempty,oneof=0 5 10"`
 }
 
 // Register handles POST /api/auth/register
@@ -550,6 +561,7 @@ func (h *AuthHandler) GetProfile(c *echo.Context) error {
 		Theme:             user.Theme.String,
 		DefaultAudioSpeed: float64(user.DefaultAudioSpeed),
 		KeepScreenAwake:   user.KeepScreenAwake,
+		AudioResumeRewind: int(user.AudioResumeRewindSeconds),
 	})
 }
 
@@ -714,6 +726,24 @@ func (h *AuthHandler) UpdateProfile(c *echo.Context) error {
 		}
 	}
 
+	// Resume-rewind preference. The form select always sends it
+	// ("0"/"5"/"10"); JSON clients that omit it decode as "" and leave
+	// the column untouched. Strict allow-list — the three choices the
+	// UI offers are the only values the API accepts.
+	if req.AudioResumeRewind != "" {
+		rewind, err := strconv.Atoi(req.AudioResumeRewind)
+		if err != nil || (rewind != 0 && rewind != 5 && rewind != 10) {
+			return c.JSON(http.StatusBadRequest, map[string]string{"error": "audio_resume_rewind_seconds must be 0, 5, or 10"})
+		}
+		err = h.db.UpdateUserAudioResumeRewind(c.Request().Context(), database.UpdateUserAudioResumeRewindParams{
+			ID:                       targetUserUUID,
+			AudioResumeRewindSeconds: int32(rewind),
+		})
+		if err != nil {
+			return c.JSON(http.StatusInternalServerError, map[string]string{"error": err.Error()})
+		}
+	}
+
 	// Keep-screen-awake preference. The form dropdown always sends it
 	// ("true"/"false"); JSON clients that omit it decode as "" and leave
 	// the column untouched.
@@ -1145,6 +1175,11 @@ func (h *AuthHandler) generateJWT(userID string) (string, error) {
 }
 
 func (h *AuthHandler) CreateDefaultCollectionsForUser(ctx context.Context, userID pgtype.UUID) error {
+	// Registration runs BEFORE the JWT exists — the request context
+	// carries no identity, and the collections RLS insert policy
+	// (2026-10-31) would reject the five system-collection seeds.
+	// Declare the new user as the actor: owner == actor, policy passes.
+	ctx = rls.WithUser(ctx, userID)
 	defaultCollections := []struct {
 		Name        string
 		Description string

internal/handlers/cards.go

@@ -43,6 +43,21 @@ type BookInfo struct {
 	// it empty rather than pay a join).
 	ViewerPercentage *float64 `json:"viewer_percentage"`
 	LibraryName      string   `json:"library_name"`
+	// Episode-collections extras (plan item 2.3): populated only by the
+	// collection-detail adapter. IsEpisode renders the episode-row
+	// variant in the grid; FeedTitle disambiguates same-named episodes;
+	// DownloadState gates the play affordance. Hidden=true means the
+	// LEFT JOIN missed under RLS (unsubscribed episode, or a
+	// family-lens hidden book) — the row renders as an "unavailable"
+	// hint instead of silently vanishing. Adapters that don't load
+	// visibility leave Hidden false (cards stay fully rendered). The
+	// four keys are NOT omitempty: every BookInfo surface emits the
+	// same stable key set (the viewer_percentage precedent —
+	// contract-pin widening, docs/contract_pins.md 2026-10-30).
+	IsEpisode     bool   `json:"is_episode"`
+	FeedTitle     string `json:"feed_title"`
+	DownloadState string `json:"download_state"`
+	Hidden        bool   `json:"hidden"`
 }
 
 // LibraryTypeMap builds libraryID→library-type-name for badge resolution
@@ -151,8 +166,7 @@ func BookCardsFromCollectionDashRows(rows []database.GetCollectionItemsForDashbo
 			ID:          uuid.UUID(mi.ID.Bytes),
 			Title:       mi.Title,
 			Author:      utils.PgTextString(mi.Author),
-			LibraryID:   mi.LibraryID,
+			LibraryID:   mi.LibraryID,			CoverPath:   mi.CoverImagePath,
-			CoverPath:   mi.CoverImagePath,
 			FormatGroup: mi.FormatGroup,
 			LibraryType: libTypes[uuid.UUID(mi.LibraryID.Bytes).String()],
 		}))
@@ -180,6 +194,39 @@ func BookCardsFromCollectionItemRows(rows []database.GetCollectionItemsRow, libT
 	return cards
 }
 
+// BookCardsFromCollectionDetailed builds cards from the
+// GetCollectionItemsDetailed rows (plan item 2.3): the collection's
+// full membership with per-row visibility. RLS-invisible rows
+// (unsubscribed episode, family-lens hidden book) become Hidden cards
+// — the grid renders them as "unavailable" hint rows; episode rows
+// carry feed title + download state for the play affordance. Unknown
+// library types resolve no badge; hidden rows need none anyway.
+func BookCardsFromCollectionDetailed(rows []database.GetCollectionItemsDetailedRow, libTypes map[string]string) []BookInfo {
+	cards := make([]BookInfo, 0, len(rows))
+	for _, r := range rows {
+		if !r.Visible {
+			cards = append(cards, BookInfo{
+				MediaItemID: uuid.UUID(r.MediaItemID.Bytes).String(),
+				Hidden:      true,
+			})
+			continue
+		}
+		isEpisode := r.Source.String == "podcast_episode"
+		card := buildBookCard(cardSource{
+			ID:          uuid.UUID(r.MediaItemID.Bytes),
+			Title:       utils.PgTextString(r.Title),
+			Author:      utils.PgTextString(r.Author),
+			CoverPath:   r.CoverImagePath,
+			FormatGroup: r.FormatGroup.String,
+		})
+		card.IsEpisode = isEpisode
+		card.FeedTitle = r.FeedTitle.String
+		card.DownloadState = r.DownloadState.String
+		cards = append(cards, card)
+	}
+	return cards
+}
+
 // BookCardsFromUnified builds cards from search-unified rows. Those
 // rows carry the library type name as a SQL column, so this adapter
 // needs no LibraryTypeMap query at all — and the viewer's own

internal/handlers/collections.go

@@ -9,7 +9,6 @@ import (
 	"errors"
 	"fmt"
 	"net/http"
-	"strconv"
 	"strings"
 	"time"
 
@@ -26,10 +25,10 @@ type CollectionHandler struct {
 	connManager       *wsync.ConnectionManager
 }
 
-func NewCollectionHandler(db *database.Queries, libraryService *services.LibraryService, connManager *wsync.ConnectionManager) *CollectionHandler {
+func NewCollectionHandler(db *database.Queries, libraryService *services.LibraryService, connManager *wsync.ConnectionManager, search ...*services.SearchService) *CollectionHandler {
 	return &CollectionHandler{
 		db:                db,
-		collectionService: services.NewCollectionService(db),
+		collectionService: services.NewCollectionService(db, search...),
 		libraryService:    libraryService,
 		connManager:       connManager,
 	}
@@ -40,17 +39,113 @@ type CreateCollectionRequest struct {
 	Description    string `form:"description" json:"description"`
 	Color          string `form:"color" json:"color"`
 	Icon           string `form:"icon" json:"icon"`
-	CollectionType  string                 `form:"collection_type" json:"collection_type"` // "static" (default) or "smart" (rule-based)
+	CollectionType string `form:"collection_type" json:"collection_type"` // "static" (default) or "smart"
-	AutoAssignRules []services.Rule        `form:"auto_assign_rules" json:"auto_assign_rules"`
+	// Custom-section builder payload (adapted 2026-10-31): the page
+	// posts the legacy rules vocabulary — rules as a JSON-encoded array
+	// (a string), a match_type select, hand-picked book ids, and the
+	// dashboard pin. The server converts rules → filter tree at write
+	// time and attaches the manual picks; storage stays tree-only.
+	AutoAssignRules json.RawMessage        `json:"auto_assign_rules"`
+	MatchType       string                 `json:"match_type"`
+	ManualBookIDs   []string               `json:"manual_book_ids"`
+	LibraryID       string                 `json:"library_id"` // selection context only — not stored
+	ShowOnDashboard bool                   `json:"show_on_dashboard"`
 	ViewSettings    map[string]interface{} `form:"view_settings" json:"view_settings"`
+	// The unified filter tree (item 3.2): when present the collection's
+	// smart membership runs through the qf engine. INVALID trees are a
+	// 400 (ParseFilterTree at the handler); a tree supersedes
+	// auto_assign_rules. Persisted as-is (any). RawMessage for
+	// round-trip safety. JSON bodies bind FilterTree; FORM bodies
+	// (the create modal's urlencoded hx-post) bind FilterTreeForm —
+	// echo's binder cannot set an any-typed field from a form string
+	// (the create-modal seed POST 400'd "invalid request" on it,
+	// 2026-10-31).
+	FilterTree any `json:"filter_tree"`
+	// FilterTreeForm mirrors filter_tree for application/x-www-form-
+	// urlencoded submissions: the raw JSON as a string, validated by
+	// the same parseRequestFilterTree path.
+	FilterTreeForm string `form:"filter_tree"`
+}
+
+// parseBuilderRules decodes the custom-section builder's rules payload:
+// the page posts JSON.stringify(rules) — a JSON string CONTAINING the
+// array (it has done so since the page was written, 2026-02-19) — while
+// JSON clients may post the array directly. Null/empty = no rules.
+func parseBuilderRules(raw json.RawMessage) ([]services.Rule, error) {
+	if len(raw) == 0 || string(raw) == "null" {
+		return nil, nil
+	}
+	var rules []services.Rule
+	if raw[0] == '"' {
+		var encoded string
+		if err := json.Unmarshal(raw, &encoded); err != nil {
+			return nil, err
+		}
+		if strings.TrimSpace(encoded) == "" {
+			return nil, nil
+		}
+		return rules, json.Unmarshal([]byte(encoded), &rules)
+	}
+	return rules, json.Unmarshal(raw, &rules)
+}
+
+// treeFilterJSON returns the request's filter tree as JSON storage
+// bytes (nil when none): object payloads marshal; strings (a legacy
+// double-encoded client, the custom-section builder's known bug class)
+// are written verbatim after object-attempt.
+func treeFilterJSON(payload any) ([]byte, error) {
+	switch v := payload.(type) {
+	case nil:
+		return nil, nil
+	case map[string]interface{}:
+		b, err := json.Marshal(v)
+		if err != nil {
+			return nil, err
+		}
+		return b, nil
+	case string:
+		return []byte(v), nil
+	default:
+		return nil, fmt.Errorf("filter_tree must be an object")
+	}
+}
+
+// parseRequestFilterTree validates the request's tree against the
+// engine grammar (ParseFilterTree) so an invalid tree is a 400, never
+// a stored 500 at render time.
+func parseRequestFilterTree(payload any) ([]byte, error) {
+	b, err := treeFilterJSON(payload)
+	if err != nil || b == nil {
+		return b, err
+	}
+	if _, err := services.ParseFilterTree(string(b)); err != nil {
+		return nil, fmt.Errorf("invalid filter_tree: %w", err)
+	}
+	return b, nil
+}
+
+// resolveTreePayload — the tree payload from either transport: JSON
+// bodies carry the object in FilterTree, form bodies (the create
+// modal's urlencoded hx-post) the raw JSON string in FilterTreeForm.
+func resolveTreePayload(jsonTree any, formTree string) any {
+	if formTree != "" {
+		return formTree
+	}
+	return jsonTree
 }
 
 type UpdateCollectionRequest struct {
 	Name string `form:"name" json:"name" validate:"required"`
+	// The unified filter tree (item 3.2): when present the collection's
+	// smart membership runs through the qf engine. INVALID trees are a
+	// 400 (ParseFilterTree at the handler). JSON bodies bind FilterTree; form bodies bind
+	// FilterTreeForm (echo's binder cannot set an any-typed field from
+	// a form string — see CreateCollectionRequest).
+	FilterTree     any                    `json:"filter_tree"`
+	FilterTreeForm string                 `form:"filter_tree"`
 	Description    string                 `form:"description" json:"description"`
 	Color          string                 `form:"color" json:"color"`
 	Icon           string                 `form:"icon" json:"icon"`
-	AutoAssignRules []services.Rule        `form:"auto_assign_rules" json:"auto_assign_rules"`
 	ViewSettings   map[string]interface{} `form:"view_settings" json:"view_settings"`
 	// A2c §3.3: playlist toggles — merged into view_settings server-side
 	// (nil = untouched). []string because the form posts a hidden "false"
@@ -94,7 +189,6 @@ type CollectionSummary struct {
 	Description string    `json:"description"`
 	Color       string    `json:"color"`
 	Icon        string    `json:"icon"`
-	AutoAssignRules json.RawMessage `json:"auto_assign_rules"`
 	CreatedAt   string    `json:"created_at"`
 	BookCount   int       `json:"book_count"`
 	IsSmart     bool      `json:"is_smart"`
@@ -138,6 +232,55 @@ func (h *CollectionHandler) CreateCollection(c *echo.Context) error {
 		return c.JSON(http.StatusBadRequest, map[string]string{"error": err.Error()})
 	}
 
+	// Custom-section builder adapter: convert the posted rules to a
+	// filter tree at WRITE time (the save-fix — the page's payload had
+	// never once bound successfully before this).
+	builderRules, rulesErr := parseBuilderRules(req.AutoAssignRules)
+	if rulesErr != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "auto_assign_rules: " + rulesErr.Error()})
+	}
+	matchMode := req.MatchType
+	if matchMode != services.RuleMatchAll && matchMode != services.RuleMatchAny {
+		matchMode = services.RuleMatchAny
+	}
+	treeJSON, treeErr := parseRequestFilterTree(resolveTreePayload(req.FilterTree, req.FilterTreeForm))
+	if treeErr != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": treeErr.Error()})
+	}
+	if treeJSON == nil && len(builderRules) > 0 {
+		tree, cerr := services.RulesToFilterTree(builderRules, matchMode)
+		if cerr != nil {
+			return c.JSON(http.StatusBadRequest, map[string]string{"error": "section rule: " + cerr.Error()})
+		}
+		b, merr := json.Marshal(tree)
+		if merr != nil {
+			return c.JSON(http.StatusInternalServerError, map[string]string{"error": merr.Error()})
+		}
+		// The strict-parse guarantee holds for converted trees too: an
+		// invalid tree is a 400 at write, never a stored 500 at render.
+		if _, verr := services.ParseFilterTree(string(b)); verr != nil {
+			return c.JSON(http.StatusBadRequest, map[string]string{"error": "section rule: " + verr.Error()})
+		}
+		treeJSON = b
+	}
+	// New podcast smart collections are BORN with the reversed listening
+	// default (2026-11-01 user request): seed chain_order into
+	// view_settings at creation — the dropdown shows "Feed order
+	// (reversed)" selected and the chain/page machinery reads the stored
+	// value exactly as if picked. Episodes-scope trees only (books have
+	// no episode chain); merge semantics never clobber a value the
+	// request itself sent; existing collections are untouched.
+	if treeJSON != nil {
+		if t, perr := services.ParseFilterTree(string(treeJSON)); perr == nil &&
+			t.FilterScope() == services.ScopeEpisodes {
+			if req.ViewSettings == nil {
+				req.ViewSettings = map[string]interface{}{}
+			}
+			if _, exists := req.ViewSettings["chain_order"]; !exists {
+				req.ViewSettings["chain_order"] = "reverse"
+			}
+		}
+	}
 	collection, err := h.collectionService.CreateCollection(
 		c.Request().Context(),
 		userUUID,
@@ -145,8 +288,9 @@ func (h *CollectionHandler) CreateCollection(c *echo.Context) error {
 		req.Description,
 		req.Color,
 		req.Icon,
-		req.AutoAssignRules,
 		req.ViewSettings,
+		req.ShowOnDashboard,
+		treeJSON,
 	)
 	if err != nil {
 		if errors.Is(err, services.ErrCollectionNameTaken) {
@@ -155,8 +299,19 @@ func (h *CollectionHandler) CreateCollection(c *echo.Context) error {
 		return c.JSON(http.StatusInternalServerError, map[string]string{"error": err.Error()})
 	}
 
+	// Manual picks: a section = rules + hand-picked books. Attach after
+	// create; unparsable ids are skipped (same leniency as AddBooks).
+	for _, bookIDStr := range req.ManualBookIDs {
+		bookID, perr := uuid.Parse(bookIDStr)
+		if perr != nil {
+			continue
+		}
+		_ = h.collectionService.AddBookToCollection(c.Request().Context(),
+			uuid.UUID(collection.ID.Bytes), bookID, userUUID)
+	}
+
 	bookCount := int32(0)
-	isSmart := req.CollectionType == "smart" || len(req.AutoAssignRules) > 0
+	isSmart := req.CollectionType == "smart" || treeJSON != nil
 	response := map[string]interface{}{
 		"id":              uuid.UUID(collection.ID.Bytes).String(),
 		"user_id":         uuid.UUID(collection.UserID.Bytes).String(),
@@ -165,7 +320,6 @@ func (h *CollectionHandler) CreateCollection(c *echo.Context) error {
 		"color":           utils.PgTextString(collection.Color),
 		"icon":            utils.PgTextString(collection.Icon),
 		"collection_type": map[bool]string{true: "smart", false: "static"}[isSmart],
-		"auto_assign_rules": collection.AutoAssignRules,
 		"view_settings":   collection.ViewSettings,
 		"book_count":      bookCount,
 		"created_at":      collection.CreatedAt.Time.String(),
@@ -197,7 +351,7 @@ func (h *CollectionHandler) GetCollections(c *echo.Context) error {
 
 	response := make([]CollectionSummary, 0, len(collections))
 	for _, col := range collections {
-		isSmart := h.collectionService.IsSmartCollection(col.AutoAssignRules)
+		isSmart := services.IsCollectionSmart(col)
 		if !includeAuto && isSmart {
 			// Legacy default for API clients (the Android app pre-dates
 			// smart collections): rule-driven collections stay opt-in so
@@ -213,12 +367,12 @@ func (h *CollectionHandler) GetCollections(c *echo.Context) error {
 			}
 		}
 		if isSmart {
-			rules := h.collectionService.ParseRules(col.AutoAssignRules)
 			visibleIDs, visErr := h.userVisibleLibraryIDs(c)
 			if visErr == nil {
-				matches, evalErr := h.collectionService.EvaluateSmartCollection(c.Request().Context(), rules, services.SmartEvalOptions{
+				matches, evalErr := h.collectionService.EvaluateSmartMembership(c.Request().Context(), col, services.SmartEvalOptions{
 					LibraryID:    libUUID,
 					VisibleIDs:   visibleIDs,
+					ViewerUserID: MustGetAuthenticatedUser(c).ID,
 				})
 				if evalErr == nil {
 					bookCount = len(matches)
@@ -240,7 +394,6 @@ func (h *CollectionHandler) GetCollections(c *echo.Context) error {
 			Description: utils.PgTextString(col.Description),
 			Color:       utils.PgTextString(col.Color),
 			Icon:        utils.PgTextString(col.Icon),
-			AutoAssignRules: col.AutoAssignRules,
 			CreatedAt:   col.CreatedAt.Time.String(),
 			BookCount:   bookCount,
 			IsSmart:     isSmart,
@@ -286,17 +439,17 @@ func (h *CollectionHandler) GetCollection(c *echo.Context) error {
 		libUUID = pgtype.UUID{Bytes: parsed, Valid: true}
 	}
 
-	isSmart := h.collectionService.IsSmartCollection(collection.AutoAssignRules)
+	isSmart := services.IsCollectionSmart(collection)
 	libTypes := LibraryTypeMap(c.Request().Context(), h.db)
 	if isSmart {
-		rules := h.collectionService.ParseRules(collection.AutoAssignRules)
 		visibleIDs, err := h.userVisibleLibraryIDs(c)
 		if err != nil {
 			return c.JSON(http.StatusInternalServerError, map[string]string{"error": err.Error()})
 		}
-		matches, evalErr := h.collectionService.EvaluateSmartCollection(c.Request().Context(), rules, services.SmartEvalOptions{
+		matches, evalErr := h.collectionService.EvaluateSmartMembership(c.Request().Context(), collection, services.SmartEvalOptions{
 			LibraryID:    libUUID,
 			VisibleIDs:   visibleIDs,
+			ViewerUserID: MustGetAuthenticatedUser(c).ID,
 		})
 		if evalErr != nil {
 			return c.JSON(http.StatusInternalServerError, map[string]string{"error": evalErr.Error()})
@@ -305,7 +458,7 @@ func (h *CollectionHandler) GetCollection(c *echo.Context) error {
 	} else if collection.IsSystemCollection.Bool && collection.QueryType.Valid && collection.QueryType.String != "" && collection.QueryType.String != "filter" {
 		user := c.Get("user").(database.Users)
 		userUUID := uuid.UUID(user.ID.Bytes)
-		dashboardSvc := services.NewDashboardService(h.db)
+		dashboardSvc := services.NewDashboardService(h.db, h.collectionService.Search())
 		sections, secErr := dashboardSvc.GetDashboardSections(c.Request().Context(), userUUID, libUUID, 1000, []string{}, []string{})
 		if secErr != nil {
 			return c.JSON(http.StatusInternalServerError, map[string]string{"error": secErr.Error()})
@@ -409,8 +562,8 @@ func (h *CollectionHandler) GetCollection(c *echo.Context) error {
 		"description":   utils.PgTextString(collection.Description),
 		"color":         utils.PgTextString(collection.Color),
 		"icon":          utils.PgTextString(collection.Icon),
-		"auto_assign_rules": json.RawMessage(collection.AutoAssignRules),
 		"view_settings": viewSettings,
+		"filter_tree":   treeEnvelope(collection.FilterTree),
 		"books":         bookList,
 		"book_count":    len(bookList),
 		"is_smart":      isSmart,
@@ -450,6 +603,31 @@ func (h *CollectionHandler) UpdateCollection(c *echo.Context) error {
 		req.ViewSettings["show_up_next"] = *v
 	}
 
+	treeJSON, treeErr := parseRequestFilterTree(resolveTreePayload(req.FilterTree, req.FilterTreeForm))
+	if treeErr != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": treeErr.Error()})
+	}
+	// Applying a tree to a collection with NO curated listening order
+	// seeds the reversed default (2026-11-01 — the creation seed's twin):
+	// a static collection becoming smart lands with the same default a
+	// born-smart one gets. Triple-guarded: the client's explicit
+	// chain_order wins, a STORED chain_order is never clobbered, and
+	// only episodes-scope trees seed (books have no episode chain).
+	if treeJSON != nil && (req.ViewSettings == nil || req.ViewSettings["chain_order"] == nil) {
+		if t, perr := services.ParseFilterTree(string(treeJSON)); perr == nil &&
+			t.FilterScope() == services.ScopeEpisodes {
+			if existing, gerr := h.collectionService.GetCollection(c.Request().Context(), collectionID); gerr == nil {
+				var stored map[string]interface{}
+				_ = json.Unmarshal(existing.ViewSettings, &stored)
+				if stored == nil || stored["chain_order"] == nil {
+					if req.ViewSettings == nil {
+						req.ViewSettings = map[string]interface{}{}
+					}
+					req.ViewSettings["chain_order"] = "reverse"
+				}
+			}
+		}
+	}
 	collection, err := h.collectionService.UpdateCollection(
 		c.Request().Context(),
 		collectionID,
@@ -457,10 +635,16 @@ func (h *CollectionHandler) UpdateCollection(c *echo.Context) error {
 		req.Description,
 		req.Color,
 		req.Icon,
-		req.AutoAssignRules,
 		req.ViewSettings,
+		treeJSON,
 	)
 	if err != nil {
+		if errors.Is(err, pgx.ErrNoRows) {
+			// Collections RLS: another user's collection is INVISIBLE —
+			// the update touched no rows. Answer 404, not a 500 that
+			// reads like a bug (2026-10-31 collections RLS).
+			return c.JSON(http.StatusNotFound, map[string]string{"error": "collection not found"})
+		}
 		return c.JSON(http.StatusInternalServerError, map[string]string{"error": err.Error()})
 	}
 
@@ -471,7 +655,6 @@ func (h *CollectionHandler) UpdateCollection(c *echo.Context) error {
 		"description":   utils.PgTextString(collection.Description),
 		"color":         utils.PgTextString(collection.Color),
 		"icon":          utils.PgTextString(collection.Icon),
-		"auto_assign_rules": json.RawMessage(collection.AutoAssignRules),
 		"view_settings": json.RawMessage(collection.ViewSettings),
 		"created_at":    collection.CreatedAt.Time.String(),
 	}
@@ -484,6 +667,99 @@ func (h *CollectionHandler) UpdateCollection(c *echo.Context) error {
 	return c.JSON(http.StatusCreated, response)
 }
 
+// ValidChainOrders — the collection chain's listening-order vocabulary
+// (plan §10 R2: a collection's playback order is CURATED on the
+// collection, not a player-side knob). Mirrors OrderEpisodes' orders;
+// "" means native (the pre-chain-order default).
+var ValidChainOrders = map[string]bool{"": true, "native": true, "newest": true, "oldest": true, "reverse": true}
+
+// CollectionChainOrder resolves a collection row's curated chain order
+// from its view_settings JSONB — "" (native) when absent/unknown. The
+// episode chain producer routes OrderEpisodes through this; the
+// view-settings PUT writes the key (SetCollectionChainOrder).
+func CollectionChainOrder(viewSettings []byte) string {
+	if len(viewSettings) == 0 {
+		return ""
+	}
+	var vs map[string]interface{}
+	if json.Unmarshal(viewSettings, &vs) != nil {
+		return ""
+	}
+	if s, ok := vs["chain_order"].(string); ok && ValidChainOrders[s] {
+		return s
+	}
+	return ""
+}
+
+// SetCollectionChainOrder — PUT /api/collections/:id/view-settings with
+// {"chain_order": "..."}: merges the key into the collection's stored
+// view_settings (the playlist toggles share that JSONB — a blind write
+// would clobber them). The episode chain producer
+// (buildSmartCollectionSequenceContext) honors it.
+func (h *CollectionHandler) SetCollectionChainOrder(c *echo.Context) error {
+	collectionID, err := uuid.Parse(c.Param("id"))
+	if err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid collection id"})
+	}
+
+	var req struct {
+		ChainOrder string `json:"chain_order"`
+	}
+	if err := c.Bind(&req); err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid request"})
+	}
+	if !ValidChainOrders[req.ChainOrder] {
+		return c.JSON(http.StatusBadRequest, map[string]string{
+			"error": "chain_order must be one of: native, newest, oldest, reverse"})
+	}
+
+	collection, err := h.collectionService.GetCollection(c.Request().Context(), collectionID)
+	if err != nil {
+		return c.JSON(http.StatusNotFound, map[string]string{"error": "collection not found"})
+	}
+	// Same owner gate as the sibling config mutation (ReorderItems) —
+	// my endpoint writes the row, so it must not let a non-owner flip
+	// someone else's chain order.
+	if u, ok := c.Get("user").(database.Users); ok {
+		owns, err := h.collectionService.UserOwnsCollection(c.Request().Context(), collectionID, uuid.UUID(u.ID.Bytes))
+		if err != nil || !owns {
+			return c.JSON(http.StatusForbidden, map[string]string{"error": "only the collection owner can change it"})
+		}
+	} else {
+		return c.JSON(http.StatusUnauthorized, map[string]string{"error": "no user in context"})
+	}
+
+	settings := map[string]interface{}{}
+	if len(collection.ViewSettings) > 0 {
+		if uerr := json.Unmarshal(collection.ViewSettings, &settings); uerr != nil {
+			return c.JSON(http.StatusInternalServerError, map[string]string{"error": "corrupt view_settings"})
+		}
+	}
+	if req.ChainOrder == "" || req.ChainOrder == "native" {
+		// Both spellings mean the default — drop the key entirely so the
+		// stored doc stays clean.
+		delete(settings, "chain_order")
+	} else {
+		settings["chain_order"] = req.ChainOrder
+	}
+	blob, err := json.Marshal(settings)
+	if err != nil {
+		return c.JSON(http.StatusInternalServerError, map[string]string{"error": err.Error()})
+	}
+	if err := h.db.UpdateCollectionViewSettings(c.Request().Context(), database.UpdateCollectionViewSettingsParams{
+		ID:           pgtype.UUID{Bytes: collectionID, Valid: true},
+		ViewSettings: blob,
+	}); err != nil {
+		return c.JSON(http.StatusInternalServerError, map[string]string{"error": err.Error()})
+	}
+
+	return c.JSON(http.StatusOK, map[string]interface{}{
+		"id":            uuid.UUID(collection.ID.Bytes).String(),
+		"chain_order":   req.ChainOrder,
+		"view_settings": settings,
+	})
+}
+
 func (h *CollectionHandler) DeleteCollection(c *echo.Context) error {
 	collectionID, err := uuid.Parse(c.Param("id"))
 	if err != nil {
@@ -851,278 +1127,25 @@ func (h *CollectionHandler) rejectSmartMutation(c *echo.Context, collectionID uu
 	if err != nil {
 		return echo.NewHTTPError(http.StatusNotFound, "collection not found")
 	}
-	if h.collectionService.IsSmartCollection(collection.AutoAssignRules) {
+	if services.IsCollectionSmart(collection) {
-		return echo.NewHTTPError(http.StatusBadRequest, "smart collections are rule-based — membership is managed by the collection's rules, not by hand")
+		return echo.NewHTTPError(http.StatusBadRequest, "smart collections are tree-based — membership is managed by the collection's filter tree, not by hand")
 	}
 	return nil
 }
 
-type AddRuleRequest struct {
+// treeEnvelope renders the stored filter tree for the envelope: nil
-	Field    string `json:"field" validate:"required"`
+// (static collections) serializes as null, trees as the object.
-	Operator string `json:"operator" validate:"required"`
+func treeEnvelope(t []byte) any {
-	Value    string `json:"value" validate:"required"`
+	if len(t) == 0 {
-	Enabled  *bool  `json:"enabled"`
+		return nil
-	Priority int    `json:"priority"`
+	}
+	var v interface{}
+	if json.Unmarshal(t, &v) == nil {
+		return v
+	}
+	return nil
 }
 
-// AddRule handles POST /api/collections/:id/rules — appends one rule to
-// the ruleset. The first rule converts a static collection to smart.
-func (h *CollectionHandler) AddRule(c *echo.Context) error {
-	collectionID, err := uuid.Parse(c.Param("id"))
-	if err != nil {
-		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid collection id"})
-	}
-
-	collection, err := h.GetCollectionData(c, collectionID)
-	if err != nil {
-		return c.JSON(http.StatusNotFound, map[string]string{"error": "collection not found"})
-	}
-	if collection.IsSystemCollection.Bool {
-		return c.JSON(http.StatusBadRequest, map[string]string{"error": "system collections cannot have rules"})
-	}
-
-	var req AddRuleRequest
-	if err := c.Bind(&req); err != nil {
-		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid request"})
-	}
-	if err := c.Validate(&req); err != nil {
-		return c.JSON(http.StatusBadRequest, map[string]string{"error": err.Error()})
-	}
-
-	updated, rule, err := h.collectionService.AddRule(c.Request().Context(), collectionID, services.Rule{
-		Field:    req.Field,
-		Operator: req.Operator,
-		Value:    req.Value,
-		Enabled:  req.Enabled,
-		Priority: req.Priority,
-	})
-	if err != nil {
-		var ruleErr *services.RuleValidationError
-		if errors.As(err, &ruleErr) {
-			return c.JSON(http.StatusBadRequest, map[string]string{"error": err.Error()})
-		}
-		return c.JSON(http.StatusInternalServerError, map[string]string{"error": err.Error()})
-	}
-
-	return c.JSON(http.StatusCreated, map[string]interface{}{
-		"collection_id": uuid.UUID(updated.ID.Bytes).String(),
-		"rule":          rule,
-		"is_smart":      h.collectionService.IsSmartCollection(updated.AutoAssignRules),
-	})
-}
-
-// UpdateRule handles PUT /api/collections/:id/rules/:ruleId — currently
-// the enabled toggle; the payload may only carry {"enabled": bool}.
-func (h *CollectionHandler) UpdateRule(c *echo.Context) error {
-	collectionID, err := uuid.Parse(c.Param("id"))
-	if err != nil {
-		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid collection id"})
-	}
-	ruleID := c.Param("ruleId")
-	if ruleID == "" {
-		return c.JSON(http.StatusBadRequest, map[string]string{"error": "rule id required"})
-	}
-
-	var req struct {
-		Enabled *bool `json:"enabled" validate:"required"`
-	}
-	if err := c.Bind(&req); err != nil {
-		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid request"})
-	}
-	if err := c.Validate(&req); err != nil {
-		return c.JSON(http.StatusBadRequest, map[string]string{"error": err.Error()})
-	}
-
-	updated, err := h.collectionService.SetRuleEnabled(c.Request().Context(), collectionID, ruleID, *req.Enabled)
-	if err != nil {
-		if strings.Contains(err.Error(), "rule not found") {
-			return c.JSON(http.StatusNotFound, map[string]string{"error": err.Error()})
-		}
-		return c.JSON(http.StatusInternalServerError, map[string]string{"error": err.Error()})
-	}
-
-	return c.JSON(http.StatusOK, map[string]interface{}{
-		"collection_id": uuid.UUID(updated.ID.Bytes).String(),
-		"is_smart":      h.collectionService.IsSmartCollection(updated.AutoAssignRules),
-	})
-}
-
-// RemoveRule handles DELETE /api/collections/:id/rules/:ruleId.
-func (h *CollectionHandler) RemoveRule(c *echo.Context) error {
-	collectionID, err := uuid.Parse(c.Param("id"))
-	if err != nil {
-		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid collection id"})
-	}
-	ruleID := c.Param("ruleId")
-	if ruleID == "" {
-		return c.JSON(http.StatusBadRequest, map[string]string{"error": "rule id required"})
-	}
-
-	updated, err := h.collectionService.RemoveRule(c.Request().Context(), collectionID, ruleID)
-	if err != nil {
-		if strings.Contains(err.Error(), "rule not found") {
-			return c.JSON(http.StatusNotFound, map[string]string{"error": err.Error()})
-		}
-		return c.JSON(http.StatusInternalServerError, map[string]string{"error": err.Error()})
-	}
-
-	return c.JSON(http.StatusOK, map[string]interface{}{
-		"collection_id": uuid.UUID(updated.ID.Bytes).String(),
-		"is_smart":      h.collectionService.IsSmartCollection(updated.AutoAssignRules),
-	})
-}
-
-type TestRulesRequest struct {
-	Rules []map[string]interface{} `json:"rules" validate:"required"`
-}
-
-type BookMatch struct {
-	MediaItemID    string `json:"media_item_id"`
-	Title          string `json:"title"`
-	Author         string `json:"author"`
-	CoverImagePath string `json:"cover_image_path"`
-	MatchReason    string `json:"match_reason"`
-}
-
-func (h *CollectionHandler) TestRules(c *echo.Context) error {
-	var req TestRulesRequest
-	if err := c.Bind(&req); err != nil {
-		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid request"})
-	}
-	if err := c.Validate(&req); err != nil {
-		return c.JSON(http.StatusBadRequest, map[string]string{"error": err.Error()})
-	}
-
-	// Candidates mirror what smart collections actually evaluate against:
-	// the requester's visible libraries and live items only — the preview
-	// must not promise matches the live render will never produce.
-	visibleIDs, err := h.userVisibleLibraryIDs(c)
-	if err != nil {
-		return c.JSON(http.StatusInternalServerError, map[string]string{"error": err.Error()})
-	}
-	var visibleArg []pgtype.UUID
-	if len(visibleIDs) > 0 {
-		visibleArg = visibleIDs
-	}
-	mediaItems, err := h.db.GetVisibleLiveMediaForRules(c.Request().Context(), database.GetVisibleLiveMediaForRulesParams{
-		LibraryID:         pgtype.UUID{},
-		VisibleLibraryIds: visibleArg,
-	})
-	if err != nil {
-		return c.JSON(http.StatusInternalServerError, map[string]string{"error": "failed to load media items"})
-	}
-
-	var matches []BookMatch
-	for _, item := range mediaItems {
-		matchReason := h.checkRulesAgainstBook(item, req.Rules)
-		if matchReason != "" {
-			author := ""
-			if item.Author.Valid {
-				author = item.Author.String
-			}
-
-			matches = append(matches, BookMatch{
-				MediaItemID:    uuid.UUID(item.ID.Bytes).String(),
-				Title:          item.Title,
-				Author:         author,
-				CoverImagePath: utils.ResolveMediaURL(item.LibraryID, item.CoverImagePath),
-				MatchReason:    matchReason,
-			})
-		}
-	}
-
-	return c.JSON(http.StatusOK, map[string]interface{}{
-		"matches": matches,
-		"total":   len(matches),
-	})
-}
-
-func (h *CollectionHandler) checkRulesAgainstBook(item database.MediaItems, rules []map[string]interface{}) string {
-	for _, rule := range rules {
-		field, _ := rule["field"].(string)
-		operator, _ := rule["operator"].(string)
-		value, _ := rule["value"].(string)
-
-		if h.evaluateRule(item, field, operator, value) {
-			return fmt.Sprintf("Matched rule: %s %s %s", field, operator, value)
-		}
-	}
-	return ""
-}
-
-func (h *CollectionHandler) evaluateRule(item database.MediaItems, field, operator, value string) bool {
-	var itemValue string
-
-	switch field {
-	case "genre":
-		if item.Genre.Valid {
-			itemValue = item.Genre.String
-		}
-	case "series":
-		if item.Series.Valid {
-			itemValue = item.Series.String
-		}
-	case "author":
-		if item.Author.Valid {
-			itemValue = item.Author.String
-		}
-	case "language":
-		if item.Language.Valid {
-			itemValue = item.Language.String
-		}
-	case "publisher":
-		if item.Publisher.Valid {
-			itemValue = item.Publisher.String
-		}
-	case "copyright_year":
-		if item.CopyrightYear.Valid {
-			itemValue = fmt.Sprintf("%d", item.CopyrightYear.Int32)
-		}
-	case "tags":
-		if len(item.Tags) > 0 {
-			itemValue = strings.Join(item.Tags, ", ")
-		}
-	}
-
-	return h.compareValues(itemValue, operator, value)
-}
-
-func (h *CollectionHandler) compareValues(itemValue, operator, ruleValue string) bool {
-	switch operator {
-	case "equals":
-		return itemValue == ruleValue
-	case "not_equals":
-		return itemValue != ruleValue
-	case "contains":
-		return strings.Contains(strings.ToLower(itemValue), strings.ToLower(ruleValue))
-	case "not_contains":
-		return !strings.Contains(strings.ToLower(itemValue), strings.ToLower(ruleValue))
-	case "starts_with":
-		return strings.HasPrefix(strings.ToLower(itemValue), strings.ToLower(ruleValue))
-	case "ends_with":
-		return strings.HasSuffix(strings.ToLower(itemValue), strings.ToLower(ruleValue))
-	case "greater_than":
-		itemNum, err1 := strconv.Atoi(itemValue)
-		ruleNum, err2 := strconv.Atoi(ruleValue)
-		if err1 != nil || err2 != nil {
-			return false
-		}
-		return itemNum > ruleNum
-	case "less_than":
-		itemNum, err1 := strconv.Atoi(itemValue)
-		ruleNum, err2 := strconv.Atoi(ruleValue)
-		if err1 != nil || err2 != nil {
-			return false
-		}
-		return itemNum < ruleNum
-	}
-	return false
-}
-
-// Bulk Operations
-
-// POST /api/collections/bulk-add-books
-// Bulk add books to multiple collections
 func (h *CollectionHandler) HandleBulkAddBooks(c *echo.Context) error {
 	user := c.Get("user").(database.Users)
 	userUUID := uuid.UUID(user.ID.Bytes)
@@ -1268,9 +1291,12 @@ func (h *CollectionHandler) PreviewCollection(c *echo.Context) error {
 		return c.JSON(http.StatusInternalServerError, map[string]string{"error": "Failed to load library items"})
 	}
 
+	// Book-scope preview (the rules page remains books-first until the
+	// filter-unification pass); no viewer progress loaded here, so
+	// played/unplayed rules preview at the no-progress default.
 	var matchedItems []database.MediaItems
 	for _, item := range allItems {
-		evaluations := h.collectionService.EvaluateRules(item, req.Rules)
+		evaluations := h.collectionService.EvaluateRules(item, 0, req.Rules)
 		for _, eval := range evaluations {
 			if eval.Matches {
 				matchedItems = append(matchedItems, item)

internal/handlers/collections_chain_order_test.go

@@ -0,0 +1,30 @@
+package handlers
+
+import "testing"
+
+// The collection chain's curated order (plan §10 R2): the resolver
+// reads view_settings.chain_order, tolerates absent/corrupt docs, and
+// refuses unknown orders (they mean native).
+func TestCollectionChainOrder(t *testing.T) {
+	cases := []struct {
+		name string
+		vs   string
+		want string
+	}{
+		{"absent doc", "", ""},
+		{"empty object", `{}`, ""},
+		{"reverse", `{"chain_order":"reverse"}`, "reverse"},
+		{"newest", `{"chain_order":"newest"}`, "newest"},
+		{"merged with toggles", `{"hide_finished":true,"chain_order":"oldest"}`, "oldest"},
+		{"unknown order means native", `{"chain_order":"shuffle"}`, ""},
+		{"non-string means native", `{"chain_order":3}`, ""},
+		{"corrupt doc means native", `not json`, ""},
+	}
+	for _, tc := range cases {
+		t.Run(tc.name, func(t *testing.T) {
+			if got := CollectionChainOrder([]byte(tc.vs)); got != tc.want {
+				t.Fatalf("CollectionChainOrder(%s) = %q, want %q", tc.vs, got, tc.want)
+			}
+		})
+	}
+}

internal/handlers/collections_rules_test.go

@@ -1,415 +0,0 @@
-package handlers
-
-import (
-	"testing"
-
-	"bookhoard/internal/database"
-
-	"github.com/jackc/pgx/v5/pgtype"
-	"github.com/stretchr/testify/assert"
-)
-
-// Test collection rule matching with various operators
-
-func TestEvaluateRule_Equals_StringMatch(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		Genre: pgtype.Text{String: "Science Fiction", Valid: true},
-	}
-
-	result := handler.evaluateRule(item, "genre", "equals", "Science Fiction")
-	assert.True(t, result)
-}
-
-func TestEvaluateRule_Equals_NoMatch(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		Genre: pgtype.Text{String: "Fantasy", Valid: true},
-	}
-
-	result := handler.evaluateRule(item, "genre", "equals", "Science Fiction")
-	assert.False(t, result)
-}
-
-func TestEvaluateRule_NotEquals_Matches(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		Genre: pgtype.Text{String: "Fantasy", Valid: true},
-	}
-
-	result := handler.evaluateRule(item, "genre", "not_equals", "Science Fiction")
-	assert.True(t, result)
-}
-
-func TestEvaluateRule_NotEquals_NoMatch(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		Genre: pgtype.Text{String: "Science Fiction", Valid: true},
-	}
-
-	result := handler.evaluateRule(item, "genre", "not_equals", "Science Fiction")
-	assert.False(t, result)
-}
-
-func TestEvaluateRule_Contains_Matches(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		Author: pgtype.Text{String: "Isaac Asimov", Valid: true},
-	}
-
-	result := handler.evaluateRule(item, "author", "contains", "asimov")
-	assert.True(t, result)
-}
-
-func TestEvaluateRule_Contains_NoMatch(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		Author: pgtype.Text{String: "Isaac Asimov", Valid: true},
-	}
-
-	result := handler.evaluateRule(item, "author", "contains", "clarke")
-	assert.False(t, result)
-}
-
-func TestEvaluateRule_NotContains_Matches(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		Author: pgtype.Text{String: "Isaac Asimov", Valid: true},
-	}
-
-	result := handler.evaluateRule(item, "author", "not_contains", "clarke")
-	assert.True(t, result)
-}
-
-func TestEvaluateRule_StartsWith_Matches(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		Series: pgtype.Text{String: "Foundation", Valid: true},
-	}
-
-	result := handler.evaluateRule(item, "series", "starts_with", "Foun")
-	assert.True(t, result)
-}
-
-func TestEvaluateRule_StartsWith_NoMatch(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		Series: pgtype.Text{String: "Foundation", Valid: true},
-	}
-
-	result := handler.evaluateRule(item, "series", "starts_with", "Robots")
-	assert.False(t, result)
-}
-
-func TestEvaluateRule_EndsWith_Matches(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		Series: pgtype.Text{String: "Foundation", Valid: true},
-	}
-
-	result := handler.evaluateRule(item, "series", "ends_with", "tion")
-	assert.True(t, result)
-}
-
-func TestEvaluateRule_GreaterThan_Matches(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		CopyrightYear: pgtype.Int4{Int32: 2021, Valid: true},
-	}
-
-	result := handler.evaluateRule(item, "copyright_year", "greater_than", "2000")
-	assert.True(t, result)
-}
-
-func TestEvaluateRule_GreaterThan_NoMatch(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		CopyrightYear: pgtype.Int4{Int32: 1999, Valid: true},
-	}
-
-	result := handler.evaluateRule(item, "copyright_year", "greater_than", "2000")
-	assert.False(t, result)
-}
-
-func TestEvaluateRule_LessThan_Matches(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		CopyrightYear: pgtype.Int4{Int32: 1999, Valid: true},
-	}
-
-	result := handler.evaluateRule(item, "copyright_year", "less_than", "2000")
-	assert.True(t, result)
-}
-
-func TestEvaluateRule_LessThan_NoMatch(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		CopyrightYear: pgtype.Int4{Int32: 2021, Valid: true},
-	}
-
-	result := handler.evaluateRule(item, "copyright_year", "less_than", "2000")
-	assert.False(t, result)
-}
-
-func TestEvaluateRule_NullField(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		Genre: pgtype.Text{Valid: false}, // NULL
-	}
-
-	result := handler.evaluateRule(item, "genre", "equals", "Science Fiction")
-	assert.False(t, result, "Should return false for NULL fields")
-}
-
-func TestCompareValues_CaseInsensitive(t *testing.T) {
-	handler := &CollectionHandler{}
-
-	result := handler.compareValues("Science Fiction", "contains", "science fiction")
-	assert.True(t, result, "Contains should be case-insensitive")
-
-	result = handler.compareValues("Science Fiction", "contains", "SCIENCE FICTION")
-	assert.True(t, result)
-}
-
-func TestCompareValues_EmptyString(t *testing.T) {
-	handler := &CollectionHandler{}
-
-	result := handler.compareValues("", "contains", "test")
-	assert.False(t, result, "Empty string should not contain anything")
-
-	result = handler.compareValues("test", "contains", "")
-	assert.True(t, result, "Non-empty string contains empty string")
-}
-
-func TestCompareValues_NumericEdgeCases(t *testing.T) {
-	handler := &CollectionHandler{}
-
-	result := handler.compareValues("0", "greater_than", "-1")
-	assert.True(t, result)
-
-	result = handler.compareValues("9999", "less_than", "10000")
-	assert.True(t, result)
-}
-
-// Test rule checking with multiple rules
-
-func TestCheckRulesAgainstBook_MatchesFirstRule(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		Genre: pgtype.Text{String: "Science Fiction", Valid: true},
-	}
-
-	rules := []map[string]interface{}{
-		{
-			"field":    "genre",
-			"operator": "equals",
-			"value":    "Science Fiction",
-		},
-		{
-			"field":    "author",
-			"operator": "contains",
-			"value":    "Asimov",
-		},
-	}
-
-	reason := handler.checkRulesAgainstBook(item, rules)
-	assert.NotEmpty(t, reason, "Should match at least one rule")
-	assert.Contains(t, reason, "genre")
-}
-
-func TestCheckRulesAgainstBook_MatchesSecondRule(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		Genre:         pgtype.Text{String: "Fantasy", Valid: true},
-		Author:        pgtype.Text{String: "Isaac Asimov", Valid: true},
-		CopyrightYear: pgtype.Int4{Int32: 2021, Valid: true},
-	}
-
-	rules := []map[string]interface{}{
-		{
-			"field":    "genre",
-			"operator": "equals",
-			"value":    "Science Fiction",
-		},
-		{
-			"field":    "author",
-			"operator": "contains",
-			"value":    "Asimov",
-		},
-	}
-
-	reason := handler.checkRulesAgainstBook(item, rules)
-	assert.NotEmpty(t, reason, "Should match second rule")
-	assert.Contains(t, reason, "author")
-}
-
-func TestCheckRulesAgainstBook_NoMatches(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		Genre:         pgtype.Text{String: "Mystery", Valid: true},
-		Author:        pgtype.Text{String: "Agatha Christie", Valid: true},
-		CopyrightYear: pgtype.Int4{Int32: 1950, Valid: true},
-	}
-
-	rules := []map[string]interface{}{
-		{
-			"field":    "genre",
-			"operator": "equals",
-			"value":    "Science Fiction",
-		},
-		{
-			"field":    "author",
-			"operator": "contains",
-			"value":    "Asimov",
-		},
-	}
-
-	reason := handler.checkRulesAgainstBook(item, rules)
-	assert.Empty(t, reason, "Should not match any rules")
-}
-
-func TestCheckRulesAgainstBook_EmptyRules(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		Genre: pgtype.Text{String: "Science Fiction", Valid: true},
-	}
-
-	rules := []map[string]interface{}{}
-
-	reason := handler.checkRulesAgainstBook(item, rules)
-	assert.Empty(t, reason, "Empty rules should return empty reason")
-}
-
-// Test complex rule scenarios
-
-func TestComplexRule_MultipleConditions(t *testing.T) {
-	handler := &CollectionHandler{}
-
-	// Book: "Foundation" by Isaac Asimov, Sci-Fi, 1951
-	item := database.MediaItems{
-		Title:         "Foundation",
-		Author:        pgtype.Text{String: "Isaac Asimov", Valid: true},
-		Genre:         pgtype.Text{String: "Science Fiction", Valid: true},
-		CopyrightYear: pgtype.Int4{Int32: 1951, Valid: true},
-		Series:        pgtype.Text{String: "Foundation", Valid: true},
-	}
-
-	tests := []struct {
-		name     string
-		field    string
-		operator string
-		value    string
-		expected bool
-	}{
-		{
-			name:     "Match genre exactly",
-			field:    "genre",
-			operator: "equals",
-			value:    "Science Fiction",
-			expected: true,
-		},
-		{
-			name:     "Match author substring",
-			field:    "author",
-			operator: "contains",
-			value:    "asimov",
-			expected: true,
-		},
-		{
-			name:     "Match year greater than",
-			field:    "copyright_year",
-			operator: "greater_than",
-			value:    "1950",
-			expected: true,
-		},
-		{
-			name:     "Match year less than",
-			field:    "copyright_year",
-			operator: "less_than",
-			value:    "2000",
-			expected: true,
-		},
-		{
-			name:     "Match series starts with",
-			field:    "series",
-			operator: "starts_with",
-			value:    "Found",
-			expected: true,
-		},
-		{
-			name:     "Match series ends with",
-			field:    "series",
-			operator: "ends_with",
-			value:    "tion",
-			expected: true,
-		},
-		{
-			name:     "No match for genre",
-			field:    "genre",
-			operator: "equals",
-			value:    "Fantasy",
-			expected: false,
-		},
-		{
-			name:     "No match for author",
-			field:    "author",
-			operator: "contains",
-			value:    "Tolkien",
-			expected: false,
-		},
-	}
-
-	for _, tt := range tests {
-		t.Run(tt.name, func(t *testing.T) {
-			result := handler.evaluateRule(item, tt.field, tt.operator, tt.value)
-			assert.Equal(t, tt.expected, result)
-		})
-	}
-}
-
-// Test edge cases and error handling
-
-func TestEvaluateRule_InvalidOperator(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		Genre: pgtype.Text{String: "Science Fiction", Valid: true},
-	}
-
-	result := handler.evaluateRule(item, "genre", "invalid_operator", "Science Fiction")
-	assert.False(t, result, "Invalid operator should return false")
-}
-
-func TestEvaluateRule_NonExistentField(t *testing.T) {
-	handler := &CollectionHandler{}
-	item := database.MediaItems{
-		Genre: pgtype.Text{String: "Science Fiction", Valid: true},
-	}
-
-	result := handler.evaluateRule(item, "nonexistent_field", "equals", "value")
-	assert.False(t, result, "Non-existent field should return false")
-}
-
-func TestCompareValues_InvalidOperator(t *testing.T) {
-	handler := &CollectionHandler{}
-
-	result := handler.compareValues("test", "invalid_op", "value")
-	assert.False(t, result, "Invalid comparison operator should return false")
-}
-
-func TestCompareValues_NumericConversion(t *testing.T) {
-	handler := &CollectionHandler{}
-
-	// Valid numeric strings
-	result := handler.compareValues("100", "greater_than", "99")
-	assert.True(t, result)
-
-	result = handler.compareValues("99", "less_than", "100")
-	assert.True(t, result)
-
-	// Invalid numeric strings (should return false)
-	result = handler.compareValues("abc", "greater_than", "100")
-	assert.False(t, result, "Non-numeric string should not compare as greater than")
-
-	result = handler.compareValues("100", "greater_than", "xyz")
-	assert.False(t, result, "Non-numeric value should not allow comparison")
-}

internal/handlers/collections_test.go

@@ -3,95 +3,12 @@ package handlers
 import (
 	"testing"
 
-	"bookhoard/internal/database"
 	"bookhoard/internal/utils"
 
 	"github.com/jackc/pgx/v5/pgtype"
 	"github.com/stretchr/testify/assert"
 )
 
-func TestCompareValues_Equals(t *testing.T) {
-	handler := &CollectionHandler{}
-
-	result := handler.compareValues("Science Fiction", "equals", "Science Fiction")
-	assert.True(t, result)
-
-	result = handler.compareValues("Science Fiction", "equals", "Fantasy")
-	assert.False(t, result)
-}
-
-func TestCompareValues_Contains(t *testing.T) {
-	handler := &CollectionHandler{}
-
-	result := handler.compareValues("Isaac Asimov", "contains", "asimov")
-	assert.True(t, result)
-
-	result = handler.compareValues("Isaac Asimov", "contains", "clarke")
-	assert.False(t, result)
-}
-
-func TestCompareValues_GreaterThan(t *testing.T) {
-	handler := &CollectionHandler{}
-
-	result := handler.compareValues("2021", "greater_than", "2000")
-	assert.True(t, result)
-
-	result = handler.compareValues("1999", "greater_than", "2000")
-	assert.False(t, result)
-}
-
-func TestCompareValues_NotEquals(t *testing.T) {
-	handler := &CollectionHandler{}
-
-	result := handler.compareValues("Science Fiction", "not_equals", "Fantasy")
-	assert.True(t, result)
-
-	result = handler.compareValues("Science Fiction", "not_equals", "Science Fiction")
-	assert.False(t, result)
-}
-
-func TestEvaluateRule_Genre(t *testing.T) {
-	item := database.MediaItems{
-		Genre: pgtype.Text{String: "Science Fiction", Valid: true},
-	}
-
-	handler := &CollectionHandler{}
-
-	result := handler.evaluateRule(item, "genre", "equals", "Science Fiction")
-	assert.True(t, result)
-
-	result = handler.evaluateRule(item, "genre", "equals", "Fantasy")
-	assert.False(t, result)
-}
-
-func TestEvaluateRule_Author(t *testing.T) {
-	item := database.MediaItems{
-		Author: pgtype.Text{String: "Isaac Asimov", Valid: true},
-	}
-
-	handler := &CollectionHandler{}
-
-	result := handler.evaluateRule(item, "author", "contains", "asimov")
-	assert.True(t, result)
-
-	result = handler.evaluateRule(item, "author", "contains", "clarke")
-	assert.False(t, result)
-}
-
-func TestEvaluateRule_CopyrightYear(t *testing.T) {
-	item := database.MediaItems{
-		CopyrightYear: pgtype.Int4{Int32: 2021, Valid: true},
-	}
-
-	handler := &CollectionHandler{}
-
-	result := handler.evaluateRule(item, "copyright_year", "greater_than", "2000")
-	assert.True(t, result)
-
-	result = handler.evaluateRule(item, "copyright_year", "less_than", "2000")
-	assert.False(t, result)
-}
-
 func TestTextToString(t *testing.T) {
 	tests := []struct {
 		name     string

internal/handlers/dashboard.go

@@ -18,10 +18,10 @@ type DashboardHandler struct {
 	dashboardService *services.DashboardService
 }
 
-func NewDashboardHandler(db *database.Queries) *DashboardHandler {
+func NewDashboardHandler(db *database.Queries, search ...*services.SearchService) *DashboardHandler {
 	return &DashboardHandler{
 		db:               db,
-		dashboardService: services.NewDashboardService(db),
+		dashboardService: services.NewDashboardService(db, search...),
 	}
 }
 

internal/handlers/media.go

@@ -243,6 +243,36 @@ func NewMediaHandler(db *database.Queries, raw database.DBTX, libraryService *se
 
 // ExecuteSearch performs search and returns results with count
 // Public wrapper for shared search logic used by both JSON and HTML endpoints
+// SearchService exposes the shared search engine for SSR surfaces
+// that render a filter tree (the collection editor and bookshelf pages
+// resolve id labels through it — 2026-11-01).
+func (h *MediaHandler) SearchService() *services.SearchService { return h.searchService }
+
+// ResolveFilterLabels — GET /api/filter/labels?field=X&ids=a,b,c names
+// id-kind filter values straight from their source tables: the client
+// path that never touches a server render (applying a SAVED filter
+// from a cold bookshelf) would otherwise fall back to the suggest
+// facet, whose rows are media-constrained and label only the ids whose
+// media matches the tree's sibling rules (2026-11-01 feed repro).
+// Visibility/RLS ride the request identity, same as the SSR embed.
+func (h *MediaHandler) ResolveFilterLabels(c *echo.Context) error {
+	user := c.Get("user").(database.Users)
+	field := c.QueryParam("field")
+	if !services.IsIDValuedField(field) {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "field is not an id-valued filter field"})
+	}
+	ids := services.ParseUUIDValues(strings.Split(c.QueryParam("ids"), ","))
+	if len(ids) == 0 {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "ids required"})
+	}
+	if len(ids) > services.MaxResolveIDs {
+		ids = ids[:services.MaxResolveIDs]
+	}
+	labels := h.SearchService().ResolveFieldLabels(
+		c.Request().Context(), user.ID, field, ids)
+	return c.JSON(http.StatusOK, map[string]interface{}{"labels": labels})
+}
+
 func (h *MediaHandler) ExecuteSearch(ctx context.Context, params services.SearchParams) ([]database.SearchMediaItemsUnifiedRow, int, error) {
 	return h.searchService.ExecuteSearch(ctx, params)
 }
@@ -1542,6 +1572,13 @@ func (mh *MediaHandler) UpdateMediaItem(c *echo.Context) error {
 		return c.JSON(http.StatusNotFound, map[string]string{"error": "media item not found"})
 	}
 
+	// Edit segregation (plan D10, task 1.8): podcast episodes are
+	// fetcher-owned (D9 — the sole writer); the generic edit box would
+	// clobber feed metadata that the next refresh overwrites anyway.
+	if existing.Source == "podcast_episode" {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "podcast episodes are managed by their feed and cannot be edited here"})
+	}
+
 	coverPath := existing.CoverImagePath.String
 	coverUploaded := false
 
@@ -1874,6 +1911,17 @@ func (mh *MediaHandler) DeleteMediaItem(c *echo.Context) error {
 		return c.JSON(http.StatusInternalServerError, map[string]string{"error": err.Error()})
 	}
 
+	// Invariant 2 (podcast plan §10, absolute form): episode rows are
+	// removed ONLY when the whole feed is removed — the feed must match
+	// its XML until then. Same segregation as the edit path (D9/D10):
+	// episodes are fetcher-owned; delete FILES via the podcast surfaces
+	// (feed page / admin podcast view) instead.
+	if media.Source == "podcast_episode" {
+		return c.JSON(http.StatusConflict, map[string]string{
+			"error": "podcast episodes are removed only by removing the whole feed — delete the episode's files instead (feed page or admin podcast view)",
+		})
+	}
+
 	// Storage-down guard (see HandleBulkDelete): refuse while the library
 	// storage is unreachable so the cascade can't destroy annotations for
 	// a file that will re-import on the next scan.
@@ -2541,6 +2589,13 @@ func ParseMediaSearchParams(c searchParamSource) services.SearchParams {
 		tree, err := services.ParseFilterTree(qf)
 		if err != nil {
 			params.FilterParseErr = err.Error()
+		} else if len(tree.Children) == 0 {
+			// A qf with no rules is not a browse query: the deep-link
+			// contract answers 400 (pinned). The empty root gained
+			// meaning only as a STORED collection seed — where the
+			// composers define it as matching nothing. As a query it
+			// would silently mean "everything", so reject at the wire.
+			params.FilterParseErr = "qf tree has no rules"
 		} else {
 			params.FilterTree = tree
 		}
@@ -3046,14 +3101,19 @@ func (mh *MediaHandler) GetAudioChapters(c *echo.Context) error {
 	// 1.0 IN the payload so the player applies it with zero extra round
 	// trips; default_playback_speed powers the dropdown's "(default)"
 	// marker. Speed is a preference, so this rides the authenticated
-	// user — anonymous readers just get 1.0.
+	// user — anonymous readers just get 1.0. The resume rewind rides
+	// the same user row (0 = exact resume).
 	playbackSpeed, defaultSpeed := 1.0, 1.0
+	resumeRewind := 0
 	if userID, ok := c.Get("user_id").(string); ok {
 		if userUUID, err := uuid.Parse(userID); err == nil {
 			userRow, err := mh.db.GetUser(c.Request().Context(), pgtype.UUID{Bytes: userUUID, Valid: true})
-			if err == nil && userRow.DefaultAudioSpeed > 0 {
+			if err == nil {
+				if userRow.DefaultAudioSpeed > 0 {
 					defaultSpeed = float64(userRow.DefaultAudioSpeed)
 				}
+				resumeRewind = int(userRow.AudioResumeRewindSeconds)
+			}
 			bookSpeed, err := mh.db.GetAudioPlaybackSpeed(c.Request().Context(), database.GetAudioPlaybackSpeedParams{
 				UserID:      pgtype.UUID{Bytes: userUUID, Valid: true},
 				MediaItemID: pgtype.UUID{Bytes: mediaUUID, Valid: true},
@@ -3075,6 +3135,7 @@ func (mh *MediaHandler) GetAudioChapters(c *echo.Context) error {
 		"tracks":                 tracks,
 		"playback_speed":         playbackSpeed,
 		"default_playback_speed": defaultSpeed,
+		"resume_rewind_seconds":  resumeRewind,
 	})
 }
 

internal/handlers/opds.go

@@ -1,5 +1,4 @@
 package handlers
-
 import (
 	"bookhoard/internal/database"
 	"bookhoard/internal/opds"
@@ -18,6 +17,24 @@ import (
 	"github.com/labstack/echo/v5"
 )
 
+// opdsVisibleLibraries is GetUserVisibleLibraries minus podcasts-type
+// libraries (plan D16): feeds are not OPDS inventory day one — they
+// surface via the /podcasts endpoints only.
+func opdsVisibleLibraries(ctx context.Context, q *database.Queries, userID pgtype.UUID) ([]database.GetUserVisibleLibrariesRow, error) {
+	libs, err := q.GetUserVisibleLibraries(ctx, userID)
+	if err != nil {
+		return nil, err
+	}
+	out := make([]database.GetUserVisibleLibrariesRow, 0, len(libs))
+	for _, lib := range libs {
+		if lib.TypeName == services.LibraryTypePodcasts {
+			continue
+		}
+		out = append(out, lib)
+	}
+	return out, nil
+}
+
 type OPDSHandler struct {
 	db                *database.Queries
 	libraryService    *services.LibraryService
@@ -209,7 +226,7 @@ func (h *OPDSHandler) GetDeviceCatalog(c *echo.Context) error {
 
 	// Get user's visible libraries
 	userID := device.UserID.Bytes
-	libraries, err := h.db.GetUserVisibleLibraries(c.Request().Context(), pgtype.UUID{Bytes: userID, Valid: true})
+	libraries, err := opdsVisibleLibraries(c.Request().Context(), h.db, pgtype.UUID{Bytes: userID, Valid: true})
 	if err != nil {
 		return c.XML(http.StatusInternalServerError, opds.NewErrorFeed("Failed to get libraries"))
 	}
@@ -318,7 +335,7 @@ func (h *OPDSHandler) serveAllBooksCatalog(c *echo.Context) error {
 
 	// Get user's visible libraries
 	userID := device.UserID.Bytes
-	libraries, err := h.db.GetUserVisibleLibraries(c.Request().Context(), pgtype.UUID{Bytes: userID, Valid: true})
+	libraries, err := opdsVisibleLibraries(c.Request().Context(), h.db, pgtype.UUID{Bytes: userID, Valid: true})
 	if err != nil {
 		return c.XML(http.StatusInternalServerError, opds.NewErrorFeed("Failed to get libraries"))
 	}
@@ -570,7 +587,7 @@ func (h *OPDSHandler) GetLibraryCatalog(c *echo.Context) error {
 
 	// The library must be visible to the device owner.
 	userID := device.UserID.Bytes
-	libraries, err := h.db.GetUserVisibleLibraries(c.Request().Context(), pgtype.UUID{Bytes: userID, Valid: true})
+	libraries, err := opdsVisibleLibraries(c.Request().Context(), h.db, pgtype.UUID{Bytes: userID, Valid: true})
 	if err != nil {
 		return c.XML(http.StatusInternalServerError, opds.NewErrorFeed("Failed to get libraries"))
 	}
@@ -677,7 +694,7 @@ func (h *OPDSHandler) SearchDeviceCatalog(c *echo.Context) error {
 
 	// Get user's visible libraries
 	userID := device.UserID.Bytes
-	libraries, err := h.db.GetUserVisibleLibraries(c.Request().Context(), pgtype.UUID{Bytes: userID, Valid: true})
+	libraries, err := opdsVisibleLibraries(c.Request().Context(), h.db, pgtype.UUID{Bytes: userID, Valid: true})
 	if err != nil {
 		return c.XML(http.StatusInternalServerError, opds.NewErrorFeed("Failed to get libraries"))
 	}
@@ -818,7 +835,7 @@ func (h *OPDSHandler) DownloadBook(c *echo.Context) error {
 
 	// Get user's visible libraries
 	userID := device.UserID.Bytes
-	libraries, err := h.db.GetUserVisibleLibraries(c.Request().Context(), pgtype.UUID{Bytes: userID, Valid: true})
+	libraries, err := opdsVisibleLibraries(c.Request().Context(), h.db, pgtype.UUID{Bytes: userID, Valid: true})
 	if err != nil {
 		return c.JSON(http.StatusInternalServerError, map[string]string{"error": "failed to get libraries"})
 	}
@@ -974,7 +991,7 @@ func (h *OPDSHandler) GetCoverImage(c *echo.Context) error {
 
 	// Get user's visible libraries
 	userID := device.UserID.Bytes
-	libraries, err := h.db.GetUserVisibleLibraries(c.Request().Context(), pgtype.UUID{Bytes: userID, Valid: true})
+	libraries, err := opdsVisibleLibraries(c.Request().Context(), h.db, pgtype.UUID{Bytes: userID, Valid: true})
 	if err != nil {
 		return c.JSON(http.StatusInternalServerError, map[string]string{"error": "failed to get libraries"})
 	}
@@ -1115,7 +1132,7 @@ func (h *OPDSHandler) ListFormats(c *echo.Context) error {
 
 	// Get user's visible libraries
 	userID := device.UserID.Bytes
-	libraries, err := h.db.GetUserVisibleLibraries(c.Request().Context(), pgtype.UUID{Bytes: userID, Valid: true})
+	libraries, err := opdsVisibleLibraries(c.Request().Context(), h.db, pgtype.UUID{Bytes: userID, Valid: true})
 	if err != nil {
 		return c.JSON(http.StatusInternalServerError, map[string]string{"error": "failed to get libraries"})
 	}

internal/handlers/podcast.go

@@ -0,0 +1,758 @@
+// Podcast handlers (plan §8): JSON API over the RLS-governed podcast
+// surfaces. Handlers parse requests and call services; visibility is
+// the engine's job (D20) — Go checks only explicit-argument
+// authorization (target-library rules, admin role) and subscription
+// lookups (podcast_subscriptions is ungoverned BY DESIGN: it IS the
+// read policy's lookup table).
+package handlers
+
+import (
+	"context"
+	"encoding/json"
+	"log"
+	"sort"
+
+	"bookhoard/internal/database"
+	"bookhoard/internal/services"
+	"net/http"
+	"strings"
+
+	"github.com/jackc/pgx/v5/pgtype"
+	"github.com/labstack/echo/v5"
+)
+
+type PodcastHandler struct {
+	db       *database.Queries
+	podcasts *services.PodcastService
+	worker   *services.PodcastDownloadWorker
+
+	// episodeListRenderer renders the refreshed episode-list partial
+	// for htmx action responses (download/refresh/order/policy).
+	// Injected by the router at wiring time — the view assembly lives
+	// with the other templ view code (handlers must not import
+	// templates: templates imports handlers for MediaDetail). When nil
+	// (tests, API-only), htmx callers fall back to JSON + HX-Trigger.
+	episodeListRenderer EpisodeListRenderer
+}
+
+// EpisodeListRenderer renders the post-action episode list for an htmx
+// request. includePolicySelect additionally out-of-band-swaps the
+// auto-download select (Download All flips the caller's policy — the
+// dropdown must reflect it immediately, issue #3);
+// includeOrderSelect likewise rides along for the order endpoint (the
+// "Reverse feed order" button flips to reverse — the dropdown must
+// mirror the server state, §0 item 3).
+type EpisodeListRenderer func(c *echo.Context, feedID string, includePolicySelect, includeOrderSelect bool) error
+
+// SetEpisodeListRenderer wires the renderer (router-only; product wiring).
+func (h *PodcastHandler) SetEpisodeListRenderer(r EpisodeListRenderer) {
+	h.episodeListRenderer = r
+}
+
+// hxEpisodeList answers an htmx action with the refreshed partial when
+// a renderer is wired; otherwise falls back to 204 + HX-Trigger.
+func (h *PodcastHandler) hxEpisodeList(c *echo.Context, feedID string, includePolicySelect, includeOrderSelect bool) error {
+	if h.episodeListRenderer != nil {
+		return h.episodeListRenderer(c, feedID, includePolicySelect, includeOrderSelect)
+	}
+	return hxAccepted(c, "")
+}
+
+// AdminDeleteFeedFiles POST /api/podcasts/feeds/:id/delete-files —
+// admin-only (§0 item 8): delete every downloaded episode file of the
+// feed; rows reset to pending. HX answers HX-Refresh (the admin table
+// re-renders with fresh aggregates).
+func (h *PodcastHandler) AdminDeleteFeedFiles(c *echo.Context) error {
+	user := MustGetAuthenticatedUser(c)
+	if user.Role != "admin" {
+		return c.JSON(http.StatusForbidden, map[string]string{"error": "admin only"})
+	}
+	feedID, err := parseUUID(c.Param("id"))
+	if err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid feed id"})
+	}
+	// media_items writes run under the first-admin identity (invariant
+	// 12 — same contract as the fetcher/worker).
+	actx, err := h.worker.AdminCtx(c.Request().Context())
+	if err != nil {
+		return podcastErrorResponse(c, err)
+	}
+	reset, err := h.podcasts.AdminDeleteFeedFiles(actx, feedID)
+	if err != nil {
+		return podcastErrorResponse(c, err)
+	}
+	log.Printf("[PODCAST] admin: deleted files for %d episode(s) of feed %s (user %s)", reset, feedID, user.Username)
+	if isHXRequest(c) {
+		c.Response().Header().Set("HX-Refresh", "true")
+		return c.JSON(http.StatusOK, map[string]interface{}{"reset_episodes": reset})
+	}
+	return c.JSON(http.StatusOK, map[string]interface{}{"reset_episodes": reset})
+}
+
+// AdminDeleteEpisodeFile POST /api/podcasts/episodes/:id/delete-file —
+// admin-only (§0 item 8): ONE episode's file deleted, row reset to
+// pending. HX answers with the refreshed episode list.
+func (h *PodcastHandler) AdminDeleteEpisodeFile(c *echo.Context) error {
+	user := MustGetAuthenticatedUser(c)
+	if user.Role != "admin" {
+		return c.JSON(http.StatusForbidden, map[string]string{"error": "admin only"})
+	}
+	episodeID, err := parseUUID(c.Param("eid"))
+	if err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid episode id"})
+	}
+	actx, err := h.worker.AdminCtx(c.Request().Context())
+	if err != nil {
+		return podcastErrorResponse(c, err)
+	}
+	if err := h.podcasts.AdminDeleteEpisodeFile(actx, episodeID); err != nil {
+		return podcastErrorResponse(c, err)
+	}
+	log.Printf("[PODCAST] admin: deleted file of episode %s (user %s)", episodeID, user.Username)
+	if isHXRequest(c) {
+		// Re-render the episode list: the row needs its fresh state. The
+		// episode's feed id comes from the row we just reset.
+		if ep, err := h.db.GetPodcastEpisode(actx, episodeID); err == nil && ep.SourceFeedID.Valid {
+			return h.hxEpisodeList(c, ep.SourceFeedID.String(), false, false)
+		}
+		return hxAccepted(c, "")
+	}
+	return c.JSON(http.StatusOK, map[string]string{"message": "file deleted"})
+}
+
+// AdminDeleteEpisodeFilesBulk POST /api/podcasts/feeds/:id/episodes/delete-files —
+// admin-only (§0 item 8): multi-select file deletion. Body carries the
+// CHECKED boxes' values (form-encoded, repeated episode-file-select
+// keys — hx-include gathers them; parsed directly, never bound: the
+// repeated-key wire is another echo-bind gotcha). Each id is validated
+// to belong to THIS feed; rows reset to pending, files removed.
+func (h *PodcastHandler) AdminDeleteEpisodeFilesBulk(c *echo.Context) error {
+	user := MustGetAuthenticatedUser(c)
+	if user.Role != "admin" {
+		return c.JSON(http.StatusForbidden, map[string]string{"error": "admin only"})
+	}
+	feedID, err := parseUUID(c.Param("id"))
+	if err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid feed id"})
+	}
+	if err := c.Request().ParseForm(); err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid request"})
+	}
+	rawIDs := c.Request().PostForm["episode-file-select"]
+	if len(rawIDs) == 0 {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "no episodes selected"})
+	}
+	actx, err := h.worker.AdminCtx(c.Request().Context())
+	if err != nil {
+		return podcastErrorResponse(c, err)
+	}
+
+	deleted := 0
+	for _, raw := range rawIDs {
+		episodeID, perr := parseUUID(raw)
+		if perr != nil {
+			continue
+		}
+		// The checkbox list only renders this feed's rows, but the wire
+		// is user-controlled: a forged id for a foreign feed's episode
+		// is refused (feed-match guard), not silently applied.
+		ep, gerr := h.db.GetPodcastEpisode(actx, episodeID)
+		if gerr != nil || ep.SourceFeedID != feedID {
+			continue
+		}
+		if derr := h.podcasts.AdminDeleteEpisodeFile(actx, episodeID); derr == nil {
+			deleted++
+		}
+	}
+	log.Printf("[PODCAST] admin: bulk-deleted files for %d/%d episode(s) of feed %s (user %s)",
+		deleted, len(rawIDs), feedID, user.Username)
+	if isHXRequest(c) {
+		return h.hxEpisodeList(c, feedID.String(), false, false)
+	}
+	return c.JSON(http.StatusOK, map[string]interface{}{"deleted": deleted})
+}
+
+// flipSubscriptionToAll implements issue #3: Download All (button or
+// add-feed checkbox) sets the CALLER's subscription policy to 'all' —
+// per-user, never global (files are shared D6, policies are not). The
+// dropdown reflects it via the renderer's oob swap; the user can turn
+// it down afterwards.
+func (h *PodcastHandler) flipSubscriptionToAll(ctx context.Context, userID, feedID pgtype.UUID) {
+	sub, err := h.db.GetPodcastSubscription(ctx, database.GetPodcastSubscriptionParams{
+		UserID: userID, FeedID: feedID,
+	})
+	if err != nil || sub.AutoDownload == "all" {
+		return // not subscribed (admin pass-through) or already all
+	}
+	_, _ = h.db.UpdatePodcastSubscription(ctx, database.UpdatePodcastSubscriptionParams{
+		UserID:            userID,
+		FeedID:            feedID,
+		DisplayName:       sub.DisplayName,
+		AutoDownload:      "all",
+		AutoDownloadCount: pgtype.Int4{}, // all needs no count
+	})
+	log.Printf("[PODCAST] flip: user %s → policy 'all' for feed %s (Download All / add checkbox)", userID, feedID)
+}
+
+func NewPodcastHandler(db *database.Queries, svc *services.PodcastService, worker *services.PodcastDownloadWorker) *PodcastHandler {
+	return &PodcastHandler{db: db, podcasts: svc, worker: worker}
+}
+
+// ---- request/response shapes ----
+
+type AddPodcastFeedRequest struct {
+	LibraryID   string   `json:"library_id" form:"library_id" validate:"required"`
+	URL         string   `json:"url" form:"url" validate:"required"`
+	DownloadAll FlexBool `json:"download_all" form:"download_all"`
+	// DownloadOrder: the initial listening order for THIS feed, applied
+	// only with DownloadAll (§0 item 5c) — it both seeds the queue in
+	// listening order and sets the caller's order override. ""/"feed" =
+	// no override row.
+	DownloadOrder string `json:"download_order,omitempty" form:"download_order"`
+}
+
+// FlexBool binds checkbox-style booleans on every wire. echo bool-binds
+// via strconv.ParseBool, which accepts "true"/"1" but REJECTS the
+// browser-native checkbox value "on" — and a bind failure is fatal for
+// the WHOLE request (the add modal would 400 "invalid request" on any
+// form variant that posts "on"). §0 item 2 (2026-10-30).
+//
+// A checkbox only contributes its field when CHECKED, so for form
+// posts presence really is the signal — JSON's native bool binding
+// ("false"→false) is unaffected. Absence on either wire = false.
+type FlexBool bool
+
+func (b *FlexBool) UnmarshalParam(v string) error {
+	switch strings.ToLower(strings.TrimSpace(v)) {
+	case "false", "0", "off", "n", "no":
+		*b = false
+	default:
+		*b = true // "true", "on", "yes", "checked", "1" — or any truthy variant
+	}
+	return nil
+}
+
+type UpdatePodcastSubscriptionRequest struct {
+	DisplayName       *string `json:"display_name" form:"display_name"`
+	AutoDownload      *string `json:"auto_download" form:"auto_download" validate:"omitempty,oneof=off all newest"`
+	AutoDownloadCount *int    `json:"auto_download_count" form:"auto_download_count"`
+}
+
+type DownloadEpisodesRequest struct {
+	EpisodeIDs []string `json:"episode_ids"`
+}
+
+// hxAccepted answers an htmx request with 204 + the podcast:changed
+// trigger — the episode-list partial re-fetches itself (its hx-trigger
+// listens for the event; polling covers download progress). JSON
+// callers keep the JSON body. HX-Redirect/HX-Trigger headers are the
+// sanctioned Go-handler response channels (CSP discipline).
+func hxAccepted(c *echo.Context, redirect string) error {
+	if redirect != "" {
+		c.Response().Header().Set("HX-Redirect", redirect)
+		return c.NoContent(http.StatusNoContent)
+	}
+	c.Response().Header().Set("HX-Trigger", "podcast:changed")
+	return c.NoContent(http.StatusNoContent)
+}
+
+func isHXRequest(c *echo.Context) bool {
+	return c.Request().Header.Get("HX-Request") == "true"
+}
+
+// podcastErrorResponse maps service errors (incl. PodcastError statuses)
+// onto HTTP responses.
+func podcastErrorResponse(c *echo.Context, err error) error {
+	if pe, ok := err.(*services.PodcastError); ok {
+		return c.JSON(pe.Status, map[string]string{"error": pe.Message})
+	}
+	return c.JSON(http.StatusInternalServerError, map[string]string{"error": err.Error()})
+}
+
+// podcastFeedSummaries is the list payload: subscription + feed metadata
+// + the facet values the /podcasts page filters on (computed here —
+// feed counts are tens, not thousands; no facet SQL needed).
+type podcastFeedSummary struct {
+	Feed
+	SubscribedAt  string `json:"subscribed_at"`
+	EpisodeTotal  int    `json:"episode_total"`
+	DownloadedN   int    `json:"episode_downloaded"`
+	UndownloadedN int    `json:"episode_undownloaded"`
+	NewEpisodes   bool   `json:"new_episodes"`
+}
+
+// AddFeed POST /api/podcasts/feeds
+func (h *PodcastHandler) AddFeed(c *echo.Context) error {
+	user := MustGetAuthenticatedUser(c)
+	var req AddPodcastFeedRequest
+	if err := c.Bind(&req); err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid request"})
+	}
+	if err := c.Validate(&req); err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": err.Error()})
+	}
+	libraryID, err := parseUUID(req.LibraryID)
+	if err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid library id"})
+	}
+
+	res, err := h.podcasts.AddFeed(c.Request().Context(), user.ID, user.Role == "admin", libraryID, req.URL)
+	if err != nil {
+		return podcastErrorResponse(c, err)
+	}
+	if req.DownloadAll && res.Feed != nil {
+		// §0 item 5c: the add modal's order dropdown is the feed's
+		// INITIAL listening order — it IS the download order (the queue
+		// seeds in that order below). 'feed'/'native' = no override row.
+		order := "native"
+		switch req.DownloadOrder {
+		case "reverse", "oldest", "newest":
+			order = req.DownloadOrder
+			payload, _ := json.Marshal(map[string]string{"order": order})
+			if err := h.db.UpsertUserSettingByKey(c.Request().Context(), database.UpsertUserSettingByKeyParams{
+				UserID: user.ID, SettingKey: podcastOrderSettingKey(res.Feed.ID.String()), SettingValue: payload,
+			}); err != nil {
+				// subscription + queue stand; the dropdown just won't
+				// preselect it (surfaced on the next order set).
+				_ = err
+			}
+		}
+		if _, err := h.worker.QueueFeedDownloadsAs(c.Request().Context(), user.ID, res.Feed.ID); err != nil {
+			// subscription stands; download-all failure surfaces via the
+			// episode list states.
+		}
+		// Issue #3: the checkbox means "everything, now AND going
+		// forward" — reflect it in the caller's subscription policy.
+		h.flipSubscriptionToAll(c.Request().Context(), user.ID, res.Feed.ID)
+	}
+	if isHXRequest(c) {
+		// Post-redirect-get: land on the new feed's page. Fetch errors
+		// (subscription stands) surface via the flash query param.
+		if res.FeedError() != "" {
+			return hxAccepted(c, "/podcasts/"+uuidString(res.Feed.ID)+"?fetch_error=1")
+		}
+		return hxAccepted(c, "/podcasts/"+uuidString(res.Feed.ID))
+	}
+	return c.JSON(http.StatusCreated, res)
+}
+
+// ListFeeds GET /api/podcasts/feeds — the caller's subscriptions.
+func (h *PodcastHandler) ListFeeds(c *echo.Context) error {
+	user := MustGetAuthenticatedUser(c)
+	ctx := c.Request().Context()
+
+	feeds, err := h.db.ListMyPodcastFeeds(ctx, user.ID)
+	if err != nil {
+		return podcastErrorResponse(c, err)
+	}
+
+	out := make([]podcastFeedSummary, 0, len(feeds))
+	for i := range feeds {
+		row := &feeds[i]
+		total, _ := h.db.CountPodcastEpisodesForFeed(ctx, row.ID)
+		downloaded, _ := h.db.CountDownloadedEpisodes(ctx, row.ID)
+		undownloaded, _ := h.db.CountUndownloadedEpisodes(ctx, row.ID)
+		out = append(out, podcastFeedSummary{
+			Feed: Feed{
+				ID: uuidString(row.ID), LibraryID: uuidString(row.LibraryID),
+				FeedUrl: row.FeedUrl, PodcastGuid: row.PodcastGuid.String,
+				Title: row.Title, Description: row.Description.String,
+				Host: row.Host.String, Publisher: row.Publisher.String,
+				Language: row.Language.String, Tags: row.Tags,
+				Explicit: row.Explicit, ArtworkPath: row.ArtworkPath.String,
+				FeedType: row.FeedType.String, Medium: row.Medium.String,
+				IsComplete: row.IsComplete, LastFetchedAt: timeString(row.LastFetchedAt),
+			},
+			SubscribedAt:  timeString(row.SubscribedAt),
+			EpisodeTotal:  int(total),
+			DownloadedN:   int(downloaded),
+			UndownloadedN: int(undownloaded),
+			// "New episodes" (D18): the feed released something the user
+			// has not pulled — undownloaded rows exist beyond the last
+			// downloaded position.
+			NewEpisodes: undownloaded > 0,
+		})
+	}
+	return c.JSON(http.StatusOK, map[string]interface{}{
+		"feeds":  out,
+		"facets": facetValuesFor(out),
+	})
+}
+
+// GetFeed GET /api/podcasts/feeds/:id — detail: metadata, episodes with
+// download state, gap summary. RLS answers 404 for non-subscribers
+// (existence leak) — no Go-level visibility plumbing.
+func (h *PodcastHandler) GetFeed(c *echo.Context) error {
+	user := MustGetAuthenticatedUser(c)
+	feedID, err := parseUUID(c.Param("id"))
+	if err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid feed id"})
+	}
+	ctx := c.Request().Context()
+
+	feed, err := h.db.GetPodcastFeed(ctx, feedID)
+	if err != nil {
+		return c.JSON(http.StatusNotFound, map[string]string{"error": "feed not found"})
+	}
+	sub, subFound := h.db.GetPodcastSubscription(ctx, database.GetPodcastSubscriptionParams{UserID: user.ID, FeedID: feedID})
+	eps, err := h.db.ListPodcastEpisodes(ctx, feedID)
+	if err != nil {
+		return podcastErrorResponse(c, err)
+	}
+
+	episodeViews := make([]PodcastEpisodeView, 0, len(eps))
+	for i := range eps {
+		episodeViews = append(episodeViews, episodeView(&eps[i]))
+	}
+	return c.JSON(http.StatusOK, map[string]interface{}{
+		"feed":               feedWithSubscription(&feed, sub, subFound == nil),
+		"episodes":           episodeViews,
+		"gap_count":          countGaps(episodeViews),
+		"undownloaded":       len(episodeViews) - countDownloaded(episodeViews),
+		"has_library_access": true, // RLS already gated the row; kept for client convenience
+	})
+}
+
+// UpdateSubscription PUT /api/podcasts/feeds/:id/subscription — the
+// only user-writable podcast state (D9): display name + auto-download.
+func (h *PodcastHandler) UpdateSubscription(c *echo.Context) error {
+	user := MustGetAuthenticatedUser(c)
+	feedID, err := parseUUID(c.Param("id"))
+	if err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid feed id"})
+	}
+	var req UpdatePodcastSubscriptionRequest
+	if err := c.Bind(&req); err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid request"})
+	}
+
+	ctx := c.Request().Context()
+	sub, err := h.db.GetPodcastSubscription(ctx, database.GetPodcastSubscriptionParams{UserID: user.ID, FeedID: feedID})
+	if err != nil {
+		// Not subscribed (or feed invisible): 404 semantics.
+		return c.JSON(http.StatusNotFound, map[string]string{"error": "not subscribed"})
+	}
+
+	displayName := sub.DisplayName
+	if req.DisplayName != nil {
+		displayName = pgtypeText(*req.DisplayName)
+	}
+	autoDownload := sub.AutoDownload
+	if req.AutoDownload != nil {
+		autoDownload = *req.AutoDownload
+	}
+	count := sub.AutoDownloadCount
+	if req.AutoDownloadCount != nil {
+		if *req.AutoDownloadCount < 0 || *req.AutoDownloadCount > 100 {
+			return c.JSON(http.StatusBadRequest, map[string]string{"error": "auto_download_count out of range"})
+		}
+		count = pgtype.Int4{Int32: int32(*req.AutoDownloadCount), Valid: *req.AutoDownloadCount > 0}
+	}
+	if autoDownload == "newest" && !count.Valid {
+		count = pgtype.Int4{Int32: 3, Valid: true} // sane default for newest-N
+	}
+
+	updated, err := h.db.UpdatePodcastSubscription(ctx, database.UpdatePodcastSubscriptionParams{
+		UserID:            user.ID,
+		FeedID:            feedID,
+		DisplayName:       displayName,
+		AutoDownload:      autoDownload,
+		AutoDownloadCount: count,
+	})
+	if err != nil {
+		return podcastErrorResponse(c, err)
+	}
+	if isHXRequest(c) {
+		// Policy changes feed the auto-download union on the next poll.
+		_, _ = h.worker.ApplyAutoPolicies(ctx, feedID)
+		return h.hxEpisodeList(c, feedID.String(), true, false)
+	}
+	return c.JSON(http.StatusOK, updated)
+}
+
+// DeleteSubscription DELETE /api/podcasts/feeds/:id/subscription —
+// unsubscribe NEVER deletes files/episodes (D6; last unsubscribe
+// retains them so re-subscribing must not re-download).
+func (h *PodcastHandler) DeleteSubscription(c *echo.Context) error {
+	user := MustGetAuthenticatedUser(c)
+	feedID, err := parseUUID(c.Param("id"))
+	if err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid feed id"})
+	}
+	rows, err := h.db.DeletePodcastSubscription(c.Request().Context(), database.DeletePodcastSubscriptionParams{UserID: user.ID, FeedID: feedID})
+	if err != nil {
+		return podcastErrorResponse(c, err)
+	}
+	if rows == 0 {
+		return c.JSON(http.StatusNotFound, map[string]string{"error": "not subscribed"})
+	}
+	if isHXRequest(c) {
+		return hxAccepted(c, "/podcasts")
+	}
+	return c.JSON(http.StatusOK, map[string]string{"message": "unsubscribed (files retained)"})
+}
+
+// RefreshFeed POST /api/podcasts/feeds/:id/refresh — manual poll.
+func (h *PodcastHandler) RefreshFeed(c *echo.Context) error {
+	user := MustGetAuthenticatedUser(c)
+	_ = user
+	feedID, err := parseUUID(c.Param("id"))
+	if err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid feed id"})
+	}
+	if err := h.podcasts.RefreshFeed(c.Request().Context(), feedID); err != nil {
+		return podcastErrorResponse(c, err)
+	}
+	if isHXRequest(c) {
+		return h.hxEpisodeList(c, feedID.String(), false, false)
+	}
+	return c.JSON(http.StatusOK, map[string]string{"message": "refreshed"})
+}
+
+// DownloadEpisodes POST /api/podcasts/feeds/:id/episodes/:eid/download
+// (+ bulk via episode_ids[]; Download All = all non-downloaded). Retry
+// is the same endpoint (idempotent state machine, plan §7).
+func (h *PodcastHandler) DownloadEpisodes(c *echo.Context) error {
+	user := MustGetAuthenticatedUser(c)
+	_ = user
+	feedID, err := parseUUID(c.Param("id"))
+	if err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid feed id"})
+	}
+	ctx := c.Request().Context()
+
+	// The feed must be in the caller's scope; RLS answers 404.
+	if _, err := h.db.GetPodcastFeed(ctx, feedID); err != nil {
+		return c.JSON(http.StatusNotFound, map[string]string{"error": "feed not found"})
+	}
+
+	// Bulk form?
+	var bulk DownloadEpisodesRequest
+	_ = c.Bind(&bulk)
+	if len(bulk.EpisodeIDs) > 0 {
+		queued := 0
+		for _, raw := range bulk.EpisodeIDs {
+			epID, err := parseUUID(raw)
+			if err != nil {
+				continue
+			}
+			if err := h.worker.QueueEpisode(ctx, epID, feedID); err == nil {
+				queued++
+			}
+		}
+		return c.JSON(http.StatusAccepted, map[string]interface{}{"queued": queued})
+	}
+
+	// Single-episode path (:eid).
+	eidRaw := c.Param("eid")
+	if eidRaw == "all" {
+		// Download All affordance: exactly the non-downloaded set (pin).
+		// Queues in the CALLER's listening order (§0 item 5b).
+		queued, err := h.worker.QueueFeedDownloadsAs(ctx, user.ID, feedID)
+		if err != nil {
+			return podcastErrorResponse(c, err)
+		}
+		if isHXRequest(c) {
+			// Download All flips the caller's policy to all (issue #3)
+			// — the oob select keeps the dropdown honest.
+			h.flipSubscriptionToAll(ctx, user.ID, feedID)
+			return h.hxEpisodeList(c, feedID.String(), true, false)
+		}
+		return c.JSON(http.StatusAccepted, map[string]interface{}{"queued": queued})
+	}
+	epID, err := parseUUID(eidRaw)
+	if err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid episode id"})
+	}
+	if err := h.worker.QueueEpisode(ctx, epID, feedID); err != nil {
+		return podcastErrorResponse(c, err)
+	}
+	if isHXRequest(c) {
+		return h.hxEpisodeList(c, feedID.String(), false, false)
+	}
+	return c.JSON(http.StatusAccepted, map[string]string{"message": "queued"})
+}
+
+// Search GET /api/podcasts/search?q= — subscription-scoped feed +
+// episode search (D14): covers un-downloaded and historical episodes
+// (rows are never deleted when a feed shrinks).
+func (h *PodcastHandler) Search(c *echo.Context) error {
+	user := MustGetAuthenticatedUser(c)
+	q := strings.TrimSpace(c.QueryParam("q"))
+	ctx := c.Request().Context()
+
+	feeds, err := h.db.SearchMyPodcastFeeds(ctx, database.SearchMyPodcastFeedsParams{UserID: user.ID, SearchQuery: q})
+	if err != nil {
+		return podcastErrorResponse(c, err)
+	}
+	eps, err := h.db.SearchMyPodcastEpisodes(ctx, q)
+	if err != nil {
+		return podcastErrorResponse(c, err)
+	}
+	episodeViews := make([]PodcastEpisodeView, 0, len(eps))
+	for i := range eps {
+		episodeViews = append(episodeViews, episodeView(&eps[i]))
+	}
+	return c.JSON(http.StatusOK, map[string]interface{}{"feeds": feeds, "episodes": episodeViews})
+}
+
+// DeleteFeed DELETE /api/podcasts/feeds/:id — admin-only removal;
+// optional ?files=false keeps the episode files on disk.
+func (h *PodcastHandler) DeleteFeed(c *echo.Context) error {
+	user := MustGetAuthenticatedUser(c)
+	if user.Role != "admin" {
+		return c.JSON(http.StatusForbidden, map[string]string{"error": "admin only"})
+	}
+	feedID, err := parseUUID(c.Param("id"))
+	if err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid feed id"})
+	}
+	ctx := c.Request().Context()
+	if _, err := h.db.GetPodcastFeed(ctx, feedID); err != nil {
+		return c.JSON(http.StatusNotFound, map[string]string{"error": "feed not found"})
+	}
+
+	keepFiles := c.QueryParam("files") == "false"
+	if !keepFiles {
+		// File deletion is best-effort before the row cascade; episodes'
+		// file_path values are relative to the download root.
+		if eps, err := h.db.ListPodcastEpisodes(ctx, feedID); err == nil {
+			if feed, err := h.db.GetPodcastFeed(ctx, feedID); err == nil {
+				h.podcasts.DeleteFeedFiles(ctx, &feed, eps)
+			}
+		}
+	}
+	if rows, err := h.db.DeletePodcastFeed(ctx, feedID); err != nil {
+		return podcastErrorResponse(c, err)
+	} else if rows == 0 {
+		return c.JSON(http.StatusNotFound, map[string]string{"error": "feed not found"})
+	}
+	if isHXRequest(c) {
+		return hxAccepted(c, "/podcasts")
+	}
+	return c.JSON(http.StatusOK, map[string]string{"message": "feed removed"})
+}
+
+// ---- Sequence (plan §8 / Phase 2: D5 order override, D7 chaining) ----
+
+// podcastOrderSettingKey is the per-user, per-feed order override key
+// in the reader_settings KV (D5).
+func podcastOrderSettingKey(feedID string) string {
+	return "podcast_order:" + feedID
+}
+
+// podcastOrderValid native (feed position) | reverse (position DESC,
+// no dates) | oldest | newest.
+func podcastOrderValid(o string) bool {
+	return services.ValidPodcastEpisodeOrder(o)
+}
+
+// GetFeedOrder reads the caller's effective order for a feed
+// (default: native — feed-native order is the D5 default).
+func (h *PodcastHandler) GetFeedOrder(c *echo.Context, feedID string) string {
+	user := MustGetAuthenticatedUser(c)
+	row, err := h.db.GetUserSettingByKey(c.Request().Context(), database.GetUserSettingByKeyParams{
+		UserID:     user.ID,
+		SettingKey: podcastOrderSettingKey(feedID),
+	})
+	if err != nil {
+		return "native"
+	}
+	var v struct {
+		Order string `json:"order"`
+	}
+	if err := json.Unmarshal(row, &v); err != nil || !podcastOrderValid(v.Order) {
+		return "native"
+	}
+	return v.Order
+}
+
+// SetFeedOrder PUT /api/podcasts/feeds/:id/order — per-user override.
+// Deleting the override (order=native) removes the KV row so the
+// default and the explicit choice are indistinguishable.
+func (h *PodcastHandler) SetFeedOrder(c *echo.Context) error {
+	user := MustGetAuthenticatedUser(c)
+	feedID, err := parseUUID(c.Param("id"))
+	if err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid feed id"})
+	}
+	var req struct {
+		Order string `json:"order" form:"order" validate:"required,oneof=native reverse oldest newest"`
+	}
+	if err := c.Bind(&req); err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid request"})
+	}
+	if err := c.Validate(&req); err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": err.Error()})
+	}
+	ctx := c.Request().Context()
+	if _, err := h.db.GetPodcastFeed(ctx, feedID); err != nil {
+		return c.JSON(http.StatusNotFound, map[string]string{"error": "feed not found"})
+	}
+	key := podcastOrderSettingKey(feedID.String())
+	if req.Order == "native" {
+		_ = h.db.DeleteUserSettingByKey(ctx, database.DeleteUserSettingByKeyParams{
+			UserID: user.ID, SettingKey: key,
+		})
+	} else {
+		payload, _ := json.Marshal(map[string]string{"order": req.Order})
+		if err := h.db.UpsertUserSettingByKey(ctx, database.UpsertUserSettingByKeyParams{
+			UserID: user.ID, SettingKey: key, SettingValue: payload,
+		}); err != nil {
+			return podcastErrorResponse(c, err)
+		}
+	}
+	if isHXRequest(c) {
+		return h.hxEpisodeList(c, feedID.String(), false, true)
+	}
+	return c.JSON(http.StatusOK, map[string]string{"order": req.Order})
+}
+
+// GetFeedSequence GET /api/podcasts/feeds/:id/sequence — the effective
+// listening order (override applied, feed-native fallback), DOWNLOADED
+// episodes only, plus the gap list (D7/D8). This payload powers both
+// the web player's chain mode and (later, §15) the media3 queue.
+func (h *PodcastHandler) GetFeedSequence(c *echo.Context) error {
+	feedID, err := parseUUID(c.Param("id"))
+	if err != nil {
+		return c.JSON(http.StatusBadRequest, map[string]string{"error": "invalid feed id"})
+	}
+	ctx := c.Request().Context()
+
+	order := h.GetFeedOrder(c, feedID.String())
+	eps, err := h.db.ListPodcastEpisodes(ctx, feedID)
+	if err != nil {
+		return podcastErrorResponse(c, err)
+	}
+
+	views := make([]PodcastEpisodeView, 0, len(eps))
+	for i := range eps {
+		views = append(views, episodeView(&eps[i]))
+	}
+	// Effective listening order (order consolidation §0 item 5a).
+	services.OrderEpisodes(views, order,
+		func(v PodcastEpisodeView) int32 { return int32(v.FeedPosition) },
+		func(v PodcastEpisodeView) string { return v.Date },
+	)
+
+	sequence := make([]PodcastEpisodeView, 0, len(views))
+	gaps := make([]PodcastEpisodeView, 0)
+	for _, v := range views {
+		if v.Downloaded {
+			sequence = append(sequence, v)
+		} else {
+			gaps = append(gaps, v)
+		}
+	}
+	return c.JSON(http.StatusOK, map[string]interface{}{
+		"feed_id":  feedID.String(),
+		"order":    order,
+		"episodes": sequence,
+		"gaps":     gaps,
+	})
+}
+
+// sortSlice: sort.Slice with a controller-friendly wrapper.
+func sortSlice[T any](s []T, less func(a, b T) bool) {
+	sort.Slice(s, func(i, j int) bool { return less(s[i], s[j]) })
+}

internal/handlers/podcast_views.go

@@ -0,0 +1,222 @@
+// Podcast view types + mappers shared by the JSON API and the SSR
+// partials (plan §8/§9). One shape per concept so the htmx partials and
+// the API can never drift.
+package handlers
+
+import (
+	"sort"
+	"time"
+
+	"bookhoard/internal/database"
+
+	"github.com/google/uuid"
+	"github.com/jackc/pgx/v5/pgtype"
+)
+
+// Feed is the API-facing feed shape (subscription fields attached when
+// the caller has one).
+type Feed struct {
+	ID           string   `json:"id"`
+	LibraryID    string   `json:"library_id"`
+	FeedUrl      string   `json:"feed_url"`
+	PodcastGuid  string   `json:"podcast_guid,omitempty"`
+	Title        string   `json:"title"`
+	Description  string   `json:"description,omitempty"`
+	Host         string   `json:"host,omitempty"`
+	Publisher    string   `json:"publisher,omitempty"`
+	Language     string   `json:"language,omitempty"`
+	Tags         []string `json:"tags,omitempty"`
+	Explicit     bool     `json:"explicit"`
+	ArtworkPath  string   `json:"artwork_path,omitempty"`
+	FeedType     string   `json:"feed_type,omitempty"`
+	Medium       string   `json:"medium,omitempty"`
+	IsComplete   bool     `json:"is_complete"`
+	LastFetchedAt string  `json:"last_fetched_at,omitempty"`
+
+	// Subscription-owned (rendered when present).
+	DisplayName       string `json:"display_name,omitempty"`
+	AutoDownload      string `json:"auto_download,omitempty"`
+	AutoDownloadCount int    `json:"auto_download_count,omitempty"`
+}
+
+// PodcastEpisodeView is the episode-list row: media facts + the
+// download lifecycle the list renders inline (D8/D19).
+type PodcastEpisodeView struct {
+	ID           string   `json:"id"`
+	FeedID       string   `json:"feed_id"`
+	Title        string   `json:"title"`
+	Subtitle     string   `json:"subtitle,omitempty"`
+	Description  string   `json:"description,omitempty"`
+	Authors      []string `json:"authors,omitempty"`
+	Hosts        []string `json:"hosts,omitempty"`
+	Guests       []string `json:"guests,omitempty"`
+	Date         string   `json:"date_published,omitempty"`
+	FeedPosition int      `json:"feed_position"`
+	Season       int      `json:"season_number,omitempty"`
+	Episode      int      `json:"episode_number,omitempty"`
+	Duration     int      `json:"declared_duration_seconds,omitempty"`
+	MimeType     string   `json:"mime_type,omitempty"`
+	CoverImage   string   `json:"cover_image_path,omitempty"`
+	EnclosureURL string   `json:"enclosure_url,omitempty"`
+
+	// Download lifecycle.
+	DownloadState    string `json:"download_state"`
+	DownloadError    string `json:"download_error,omitempty"`
+	DownloadAttempts int    `json:"download_attempts"`
+	Downloaded       bool   `json:"downloaded"`
+	FileSize         int64  `json:"file_size,omitempty"`
+
+	// Progress (from the shared reading_progress machinery).
+	ProgressPercent float64 `json:"progress_percent,omitempty"`
+}
+
+// episodeView maps a media_items row (podcast episode) onto the view.
+func episodeView(ep *database.MediaItems) PodcastEpisodeView {
+	v := PodcastEpisodeView{
+		ID:           uuidString(ep.ID),
+		FeedID:       uuidString(ep.SourceFeedID),
+		Title:        ep.Title,
+		Subtitle:     ep.Subtitle.String,
+		Description:  ep.Description.String,
+		Authors:      ep.Authors,
+		Hosts:        ep.Hosts,
+		Guests:       ep.Guests,
+		FeedPosition: int(ep.FeedPosition.Int32),
+		Season:       int(ep.SeasonNumber.Int32),
+		Episode:      int(ep.EpisodeNumber.Int32),
+		Duration:     int(ep.DeclaredDurationSeconds.Int32),
+		MimeType:     ep.MimeType.String,
+		CoverImage:   ep.CoverImagePath.String,
+		EnclosureURL: ep.EnclosureUrl.String,
+
+		DownloadState:    ep.DownloadState,
+		DownloadError:    ep.DownloadError.String,
+		DownloadAttempts: int(ep.DownloadAttempts),
+		Downloaded:       ep.DownloadState == "downloaded",
+		FileSize:         ep.FileSize.Int64,
+	}
+	if ep.DatePublished.Valid {
+		v.Date = ep.DatePublished.Time.Format("2006-01-02")
+	}
+	if !ep.FileSize.Valid {
+		v.FileSize = 0
+	}
+	return v
+}
+
+// feedWithSubscription merges the feed row + the caller's subscription
+// (empty sub → bare feed).
+func feedWithSubscription(feed *database.PodcastFeeds, sub database.PodcastSubscriptions, subFound bool) Feed {
+	f := Feed{
+		ID:            uuidString(feed.ID),
+		LibraryID:     uuidString(feed.LibraryID),
+		FeedUrl:       feed.FeedUrl,
+		PodcastGuid:   feed.PodcastGuid.String,
+		Title:         feed.Title,
+		Description:   feed.Description.String,
+		Host:          feed.Host.String,
+		Publisher:     feed.Publisher.String,
+		Language:      feed.Language.String,
+		Tags:          feed.Tags,
+		Explicit:      feed.Explicit,
+		ArtworkPath:   feed.ArtworkPath.String,
+		FeedType:      feed.FeedType.String,
+		Medium:        feed.Medium.String,
+		IsComplete:    feed.IsComplete,
+		LastFetchedAt: timeString(feed.LastFetchedAt),
+	}
+	if subFound {
+		f.DisplayName = sub.DisplayName.String
+		f.AutoDownload = sub.AutoDownload
+		f.AutoDownloadCount = int(sub.AutoDownloadCount.Int32)
+	}
+	return f
+}
+
+// countGaps counts un-downloaded episodes that have a downloaded
+// episode AFTER them in list order — the D8 "middle episodes missing"
+// definition that drives the gap banner.
+func countGaps(eps []PodcastEpisodeView) int {
+	seenDownloaded := false
+	gaps := 0
+	for i := len(eps) - 1; i >= 0; i-- {
+		if eps[i].Downloaded {
+			seenDownloaded = true
+		} else if seenDownloaded {
+			gaps++
+		}
+	}
+	return gaps
+}
+
+func countDownloaded(eps []PodcastEpisodeView) int {
+	n := 0
+	for _, e := range eps {
+		if e.Downloaded {
+			n++
+		}
+	}
+	return n
+}
+
+// facetValuesFor derives the /podcasts facet bar values (D18) from the
+// subscription list — feed counts are small; no facet SQL.
+func facetValuesFor(feeds []podcastFeedSummary) map[string][]string {
+	hosts := map[string]bool{}
+	tags := map[string]bool{}
+	langs := map[string]bool{}
+	media := map[string]bool{}
+	for _, f := range feeds {
+		if f.Host != "" {
+			hosts[f.Host] = true
+		}
+		for _, t := range f.Tags {
+			tags[t] = true
+		}
+		if f.Language != "" {
+			langs[f.Language] = true
+		}
+		if f.Medium != "" {
+			media[f.Medium] = true
+		}
+	}
+	return map[string][]string{
+		"hosts":    sortedKeys(hosts),
+		"tags":     sortedKeys(tags),
+		"language": sortedKeys(langs),
+		"medium":   sortedKeys(media),
+	}
+}
+
+func sortedKeys(set map[string]bool) []string {
+	out := make([]string, 0, len(set))
+	for k := range set {
+		out = append(out, k)
+	}
+	sort.Strings(out)
+	return out
+}
+
+// uuidString renders a pgtype.UUID (empty string when NULL).
+func uuidString(id pgtype.UUID) string {
+	if !id.Valid {
+		return ""
+	}
+	return uuid.UUID(id.Bytes).String()
+}
+
+// timeString renders a timestamp (RFC3339; empty when NULL).
+func timeString(t pgtype.Timestamptz) string {
+	if !t.Valid {
+		return ""
+	}
+	return t.Time.Format(time.RFC3339)
+}
+
+// pgtypeText (handlers-local) — service has its own.
+func pgtypeText(s string) pgtype.Text {
+	if s == "" {
+		return pgtype.Text{}
+	}
+	return pgtype.Text{String: s, Valid: true}
+}

internal/handlers/refresh_token.go

@@ -74,6 +74,23 @@ func (h *AuthHandler) RefreshAccessToken(c *echo.Context) error {
 
 // Logout handles POST /api/auth/logout
 func (h *AuthHandler) Logout(c *echo.Context) error {
+	// Clear the HttpOnly session cookie — the SSR pages authenticate
+	// with it, and JavaScript cannot clear an HttpOnly cookie by
+	// design. Without this, "logout" wiped localStorage but left the
+	// cookie session alive for up to session_duration_seconds (the
+	// fake-signout bug: users roamed every cookie-authed page). The
+	// cookie is cleared BEFORE anything else — even a malformed body
+	// must sign the browser out.
+	c.SetCookie(&http.Cookie{
+		Name:     "token",
+		Value:    "",
+		Path:     "/",
+		HttpOnly: true,
+		Secure:   secure == "true",
+		SameSite: http.SameSiteLaxMode,
+		MaxAge:   -1, // delete
+	})
+
 	var req RefreshTokenRequest
 	if err := c.Bind(&req); err != nil {
 		return c.JSON(http.StatusOK, map[string]string{"message": "logged out successfully"})

internal/handlers/scanner.go

@@ -85,6 +85,11 @@ func (h *Handler) ScanLibrary(c *echo.Context) error {
 			for _, folder := range libraryFolders {
 				if folder.FolderPath == req.FolderPath {
 					registered = true
+					// Podcast download roots are feed-managed (invariant
+					// 4): never scanner-ingestible, even scoped.
+					if folder.Purpose == services.FolderPurposePodcast {
+						return c.JSON(http.StatusBadRequest, map[string]string{"error": "podcast download roots are feed-managed and cannot be scanned"})
+					}
 					break
 				}
 			}

internal/handlers/system_settings.go

@@ -178,6 +178,21 @@ func validateSettingValue(def database.SettingDefault, value string) error {
 		if _, err := strconv.ParseBool(value); err != nil {
 			return fmt.Errorf("value must be true or false")
 		}
+	case database.SettingTypeFloat:
+		f, err := strconv.ParseFloat(value, 64)
+		if err != nil {
+			return fmt.Errorf("value must be a number")
+		}
+		if def.Min != "" {
+			if mn, err := strconv.ParseFloat(def.Min, 64); err == nil && f < mn {
+				return fmt.Errorf("value must be >= %s", def.Min)
+			}
+		}
+		if def.Max != "" {
+			if mx, err := strconv.ParseFloat(def.Max, 64); err == nil && f > mx {
+				return fmt.Errorf("value must be <= %s", def.Max)
+			}
+		}
 	case database.SettingTypeString:
 		if value == "" {
 			return fmt.Errorf("value must not be empty")

internal/router/collections.go

@@ -17,17 +17,16 @@ func registerCollectionsRoutes(cfg *Config) {
 	collections.GET("/:id/books", cfg.CollectionHandler.GetBookCollections)
 	collections.POST("/:id/books", cfg.CollectionHandler.AddBooks)
 	collections.PUT("/:id/items/order", cfg.CollectionHandler.ReorderItems)
+	// The chain's listening order lives ON the collection (plan §10 R2):
+	// merges into view_settings; the episode chain producer honors it.
+	collections.PUT("/:id/view-settings", cfg.CollectionHandler.SetCollectionChainOrder)
 	collections.DELETE("/:id/books/:bookId", cfg.CollectionHandler.RemoveBook)
 	collections.POST("/:id/books/bulk-remove", cfg.CollectionHandler.BulkRemoveBooks)
 	collections.POST("/bulk-add-books", cfg.CollectionHandler.HandleBulkAddBooks)
-	collections.POST("/test-rules", cfg.CollectionHandler.TestRules)
 	collections.POST("/preview", cfg.CollectionHandler.PreviewCollection)
 
 	// Smart collection rule CRUD (the rules builder persists through
 	// these; type follows rules presence — first rule = smart).
-	collections.POST("/:id/rules", cfg.CollectionHandler.AddRule)
-	collections.PUT("/:id/rules/:ruleId", cfg.CollectionHandler.UpdateRule)
-	collections.DELETE("/:id/rules/:ruleId", cfg.CollectionHandler.RemoveRule)
 
 	// Device shelf mapping routes
 	deviceCollections := protected.Group("/devices/:id/collections")

internal/router/frontend.go

@@ -14,6 +14,7 @@ import (
 
 	"bookhoard/internal/config"
 	"bookhoard/internal/database"
+	"bookhoard/internal/rls"
 	"bookhoard/internal/handlers"
 	"bookhoard/internal/services"
 	"bookhoard/internal/utils"
@@ -408,9 +409,19 @@ func registerFrontendRoutes(cfg *Config) {
 			}
 		}
 
+		// SSR id-label resolution for a URL-carried qf tree: the drawer's
+		// value cells show NAMES from first paint instead of raw uuids
+		// (2026-11-01 — same story as the collection editor above).
+		labelsJSON := "{}"
+		if qfRaw := c.QueryParam("qf"); qfRaw != "" {
+			if t, perr := services.ParseFilterTree(qfRaw); perr == nil {
+				labelsJSON = resolveTreeLabelsJSON(c, cfg, user.ID, t)
+			}
+		}
+
 		var buf bytes.Buffer
 		bookInfoList = handlers.MarkActiveConflicts(c.Request().Context(), cfg.Queries, pgtype.UUID{Bytes: userUUID, Valid: true}, bookInfoList)
-		err = templates.BookShelf(user, libData, libraryID, errorMsg, savedFilters, bookInfoList, limit, offset, totalCount).Render(c.Request().Context(), &buf)
+		err = templates.BookShelf(user, libData, libraryID, errorMsg, savedFilters, bookInfoList, limit, offset, totalCount, labelsJSON).Render(c.Request().Context(), &buf)
 		if err != nil {
 			return err
 		}
@@ -561,13 +572,17 @@ func registerFrontendRoutes(cfg *Config) {
 		colData := make([]templates.CollectionData, len(collections))
 		colService := cfg.CollectionHandler.CollectionService()
 		for i, col := range collections {
-			isSmart := colService.IsSmartCollection(col.AutoAssignRules)
+			isSmart := services.IsCollectionSmart(col)
 			bookCount := 0
 			if isSmart {
-				matches, evalErr := colService.EvaluateSmartCollection(c.Request().Context(),
+				matches, evalErr := colService.EvaluateSmartMembership(c.Request().Context(),
-					colService.ParseRules(col.AutoAssignRules), services.SmartEvalOptions{
+					col, services.SmartEvalOptions{
 						LibraryID:    libScope,
 						VisibleIDs:   visibleIDs,
+						ViewerUserID: func() pgtype.UUID {
+							uid, _ := uuid.Parse(user.ID)
+							return uuidToPGType(uid)
+						}(),
 					})
 				if evalErr == nil {
 					bookCount = len(matches)
@@ -695,6 +710,35 @@ func registerFrontendRoutes(cfg *Config) {
 			return renderErrorPage(c, "System collections cannot have rules", "not_allowed")
 		}
 
+		// The editor's vocabulary + prefill (item 3.4): the unified
+		// builder in the collection's scope. A stored tree renders
+		// as-is. (The legacy rules-array preload retired 2026-10-31
+		// with the rules-array storage — the tree is the only form.)
+		scope := "books"
+		treeJSON := "{}"
+		var parsedTree *services.FilterTree
+		if len(collection.FilterTree) > 0 {
+			if t, perr := services.ParseFilterTree(string(collection.FilterTree)); perr == nil {
+				scope = t.FilterScope()
+				if b, merr := json.Marshal(t); merr == nil {
+					treeJSON = string(b)
+				}
+				parsedTree = t
+			}
+		}
+		fieldsPayload, merr := json.Marshal(services.FilterFieldsPayload(scope))
+		if merr != nil {
+			fieldsPayload = []byte("[]")
+		}
+		fieldsJSON := string(fieldsPayload)
+
+		// SSR id-label resolution: the stored tree arrives as bare ids;
+		// embed the id→name map so value cells render NAMES at first
+		// paint (2026-11-01: the client's pick-time cache dies with the
+		// page, and the quiet suggest fetch can't label ids whose media
+		// doesn't match the tree's sibling rules — raw uuids on refresh).
+		labelsJSON := resolveTreeLabelsJSON(c, cfg, user.ID, parsedTree)
+
 		colData := templates.CollectionData{
 			ID:          collectionID,
 			Name:        collection.Name,
@@ -702,11 +746,11 @@ func registerFrontendRoutes(cfg *Config) {
 			Color:       utils.PgTextString(collection.Color),
 			Icon:        utils.PgTextString(collection.Icon),
 			IsSystem:    collection.IsSystemCollection.Bool,
-			IsSmart:     cfg.CollectionHandler.CollectionService().IsSmartCollection(collection.AutoAssignRules),
+			IsSmart:     services.IsCollectionSmart(collection),
 		}
 
 		var buf bytes.Buffer
-		err = templates.CollectionRules(user, colData).Render(c.Request().Context(), &buf)
+		err = templates.CollectionRules(user, colData, fieldsJSON, treeJSON, scope, labelsJSON).Render(c.Request().Context(), &buf)
 		if err != nil {
 			return err
 		}
@@ -1507,8 +1551,7 @@ func registerFrontendRoutes(cfg *Config) {
 
 	// Admin hash conflicts page: content-duplicate groups flagged during hash
 	// backfill or rescan, resolved by keeping all copies or merging into one.
-	frontendProtected.GET("/admin/hash-conflicts", handlers.AdminMiddleware(func(c *echo.Context) error {
+	frontendProtected.GET("/admin/hash-conflicts", handlers.AdminMiddleware(func(c *echo.Context) error {		user, err := getTemplateUserWithTheme(c, cfg)
-		user, err := getTemplateUserWithTheme(c, cfg)
 		if err != nil {
 			return renderErrorPage(c, "Error loading user", "user_load_error")
 		}
@@ -1562,6 +1605,62 @@ func registerFrontendRoutes(cfg *Config) {
 		return c.HTML(http.StatusOK, buf.String())
 	}))
 
+	// Admin all-podcasts view (§0 item 8): every feed from every user with
+	// maintenance aggregates (disk usage first). The aggregate read runs
+	// under the FIRST-ADMIN identity — the RLS admin escape (invariant
+	// 12, poller/fetcher contract) — so admins see feeds from libraries
+	// hidden even from them.
+	frontendProtected.GET("/admin/podcasts", handlers.AdminMiddleware(func(c *echo.Context) error {
+		user, err := getTemplateUserWithTheme(c, cfg)
+		if err != nil {
+			return renderErrorPage(c, "Error loading user", "user_load_error")
+		}
+
+		adminUUID, err := cfg.Queries.GetFirstAdmin(context.Background())
+		if err != nil {
+			return renderErrorPage(c, "Error loading podcasts", "podcasts_load_error")
+		}
+		// cfg.Queries is ScopedDB-backed: the context identity is all the
+		// aggregate read needs.
+		actx := rls.WithUser(c.Request().Context(), adminUUID)
+
+		rows, err := cfg.Queries.ListAllPodcastFeedsAdmin(actx)
+		if err != nil {
+			return renderErrorPage(c, "Error loading podcasts", "podcasts_load_error")
+		}
+
+		feeds := make([]templates.AdminPodcastFeedData, 0, len(rows))
+		var totalBytes int64
+		totalDownloaded, totalEpisodes := 0, 0
+		for _, r := range rows {
+			lastFetch := "never"
+			if r.LastFetchedAt.Valid {
+				lastFetch = r.LastFetchedAt.Time.In(utils.UserLocation(user.Timezone)).Format("Jan 2, 2006 15:04")
+			}
+			feeds = append(feeds, templates.AdminPodcastFeedData{
+				ID:                uuid.UUID(r.ID.Bytes).String(),
+				Title:             r.Title,
+				Host:              r.Host.String,
+				LibraryName:       r.LibraryName,
+				Subscribers:       int(r.SubscriberCount),
+				SubscriberNames:   r.SubscriberNames,
+				EpisodeTotal:      int(r.EpisodeTotal),
+				EpisodeDownloaded: int(r.EpisodeDownloaded),
+				DiskBytes:         r.DiskBytes,
+				LastFetched:       lastFetch,
+			})
+			totalBytes += r.DiskBytes
+			totalEpisodes += int(r.EpisodeTotal)
+			totalDownloaded += int(r.EpisodeDownloaded)
+		}
+
+		var buf bytes.Buffer
+		if err := templates.AdminPodcasts(user, feeds, len(feeds), totalDownloaded, totalEpisodes, totalBytes).Render(c.Request().Context(), &buf); err != nil {
+			return err
+		}
+		return c.HTML(http.StatusOK, buf.String())
+	}))
+
 	// Admin users page
 	frontendProtected.GET("/admin/users", handlers.AdminMiddleware(func(c *echo.Context) error {
 		user, err := getTemplateUserWithTheme(c, cfg)
@@ -1984,6 +2083,7 @@ func (cfg *Config) GetHealth(c *echo.Context) error {
 	return c.JSON(http.StatusOK, map[string]interface{}{
 		"status":   "healthy",
 		"version":  version.Version,
+		"commit":   version.Commit,
 		"database": "connected",
 		"scan": map[string]interface{}{
 			"scan_in_progress": cfg.Worker.HasActiveScans(),

internal/router/helpers.go

@@ -2,6 +2,7 @@ package router
 
 import (
 	"context"
+	"encoding/json"
 	"log"
 	"net/http"
 	"net/url"
@@ -55,6 +56,14 @@ func getTemplateUserWithTheme(c *echo.Context, cfg *Config) (templates.User, err
 		token = cookie.Value
 	}
 
+	// Sidebar /podcasts gate (plan D13): >=1 visible podcasts library.
+	// One indexed COUNT per page render — the same class of per-request
+	// lookup the live-user guard already does.
+	showPodcasts := false
+	if n, err := cfg.Queries.CountPodcastLibrariesForUser(c.Request().Context(), uuidToPGType(userUUID)); err == nil && n > 0 {
+		showPodcasts = true
+	}
+
 	return templates.User{
 		ID:                userID,
 		Email:             userEmail,
@@ -67,7 +76,9 @@ func getTemplateUserWithTheme(c *echo.Context, cfg *Config) (templates.User, err
 		LastName:          utils.PgTextString(userDB.LastName),
 		CreatedAt:         userDB.CreatedAt.Time,
 		DefaultAudioSpeed: float64(userDB.DefaultAudioSpeed),
+		AudioResumeRewind: int(userDB.AudioResumeRewindSeconds),
 		KeepScreenAwake:   userDB.KeepScreenAwake,
+		ShowPodcastsNav:   showPodcasts,
 	}, nil
 }
 
@@ -261,6 +272,15 @@ func visibleLibraryIDsFor(c *echo.Context, cfg *Config) (map[string]bool, []pgty
 // collectionDataFromRow builds the template collection payload shared by
 // the collection detail page and its books-grid partial.
 func collectionDataFromRow(collection database.Collections, collectionID string, isSmart bool, bookCount int) templates.CollectionData {
+	chainOrder := ""
+	if len(collection.ViewSettings) > 0 {
+		var vs map[string]interface{}
+		if json.Unmarshal(collection.ViewSettings, &vs) == nil {
+			if s, ok := vs["chain_order"].(string); ok {
+				chainOrder = s
+			}
+		}
+	}
 	return templates.CollectionData{
 		ID:          collectionID,
 		Name:        collection.Name,
@@ -269,6 +289,7 @@ func collectionDataFromRow(collection database.Collections, collectionID string,
 		Icon:        collection.Icon.String,
 		IsSystem:    collection.IsSystemCollection.Bool,
 		IsSmart:     isSmart,
+		ChainOrder:  chainOrder,
 		BookCount:   bookCount,
 	}
 }
@@ -277,25 +298,72 @@ func collectionDataFromRow(collection database.Collections, collectionID string,
 // detail page and the books-grid partial: smart evaluation, a dashboard
 // query section, or manual membership. Callers resolve the scope:
 // libUUID/libScope/isAll/scoped per their own resolution rules.
+// resolveTreeLabelsJSON — SSR id-label resolution for a page-rendered
+// filter tree: walks the tree's id-kind rules, resolves uuids to display
+// names through the search engine (RLS/visibility ride the request
+// identity), and returns the JSON map embedded into the builder mount.
+// Best-effort: any failure yields "{}" and the page renders with the
+// client's fallbacks intact (2026-11-01, filter_labels.go for the WHY).
+func resolveTreeLabelsJSON(c *echo.Context, cfg *Config, userID string, tree *services.FilterTree) string {
+	if tree == nil || len(tree.Children) == 0 {
+		return "{}"
+	}
+	userUUID, err := uuid.Parse(userID)
+	if err != nil {
+		return "{}"
+	}
+	labels := cfg.MediaHandler.SearchService().FilterTreeLabels(
+		c.Request().Context(), pgtype.UUID{Bytes: userUUID, Valid: true}, tree)
+	if len(labels) == 0 {
+		return "{}"
+	}
+	b, merr := json.Marshal(labels)
+	if merr != nil {
+		return "{}"
+	}
+	return string(b)
+}
+
 func (cfg *Config) collectionBooksForScope(c *echo.Context, userUUID uuid.UUID, collUUID pgtype.UUID, collection database.Collections, collectionID string, libUUID pgtype.UUID, libScope pgtype.UUID, scoped bool, visibleIDs []pgtype.UUID) (books []handlers.BookInfo, isSmart bool, errCode string) {
 	colService := cfg.CollectionHandler.CollectionService()
-	isSmart = colService.IsSmartCollection(collection.AutoAssignRules)
+	isSmart = services.IsCollectionSmart(collection)
 	libTypes := handlers.LibraryTypeMap(c.Request().Context(), cfg.Queries)
 
 	if isSmart {
 		// Smart (dynamic) collection: membership is evaluated live —
-		// visible libraries, live items only, any enabled rule matches.
+		// visible libraries, live items only, rules applied under the
-		matches, evalErr := colService.EvaluateSmartCollection(c.Request().Context(),
+		// collection's match mode (item 2.4: any|all), with the
-			colService.ParseRules(collection.AutoAssignRules), services.SmartEvalOptions{
+		// viewer's progress feeding played/unplayed rules.
+		matches, evalErr := colService.EvaluateSmartMembership(c.Request().Context(),
+			collection, services.SmartEvalOptions{
 				LibraryID:    libScope,
 				VisibleIDs:   visibleIDs,
+				ViewerUserID: pgtype.UUID{Bytes: userUUID, Valid: true},
 			})
 		if evalErr != nil {
 			return nil, isSmart, "books_load_error"
 		}
+		// The visible order IS the play order: the curated listening
+		// order (view_settings.chain_order, §10 R2) re-orders the
+		// episode cards exactly as the sequence producer will play
+		// them, so the select's effect is visible and Play launches
+		// from the card the user sees first (2026-11-01 user report:
+		// the select changed nothing on the page).
+		services.OrderEpisodeMedia(matches, handlers.CollectionChainOrder(collection.ViewSettings))
 		books = handlers.BookCardsFromMediaItems(matches, libTypes)
+		// Episode-awareness for smart matches (item 2.5): episode cards
+		// carry the play affordance (chain via ?queue=collection:).
+		// BookCardsFromMediaItems preserves match order, so a parallel
+		// walk is enough; feed titles stay empty here (batch lookup is
+		// polish, not acceptance).
+		for i := range books {
+			if i < len(matches) && matches[i].Source == "podcast_episode" {
+				books[i].IsEpisode = true
+				books[i].DownloadState = matches[i].DownloadState
+			}
+		}
 	} else if collection.IsSystemCollection.Bool && collection.QueryType.Valid && collection.QueryType.String != "" && collection.QueryType.String != "filter" {
-		dashboardSvc := services.NewDashboardService(cfg.Queries)
+		dashboardSvc := services.NewDashboardService(cfg.Queries, cfg.CollectionHandler.CollectionService().Search())
 		sections, secErr := dashboardSvc.GetDashboardSections(c.Request().Context(), userUUID, libUUID, 1000, []string{}, []string{})
 		if secErr != nil {
 			return nil, isSmart, "books_load_error"
@@ -322,11 +390,17 @@ func (cfg *Config) collectionBooksForScope(c *echo.Context, userUUID uuid.UUID,
 				books = handlers.BookCardsFromCollectionDashRows(collItems, libTypes)
 			}
 		} else {
-			collItems, collErr := cfg.Queries.GetCollectionItems(c.Request().Context(), collUUID)
+			// Episode collections (plan item 2.3): the detailed query
+			// keeps RLS-invisible rows as Hidden cards (unsubscribed
+			// episodes render an "unavailable" hint) and tags episode
+			// rows with feed title + download state. Same position
+			// ordering; the legacy INNER-JOIN shape remains on the
+			// library-scoped branch above.
+			collItems, collErr := cfg.Queries.GetCollectionItemsDetailed(c.Request().Context(), collUUID)
 			if collErr != nil {
 				collItems = nil
 			}
-			books = handlers.BookCardsFromCollectionItemRows(collItems, libTypes)
+			books = handlers.BookCardsFromCollectionDetailed(collItems, libTypes)
 		}
 	}
 	return books, isSmart, ""

internal/router/podcast.go

@@ -0,0 +1,83 @@
+package router
+
+// Podcast routes (plan §8). All under /api/podcasts with the JWT
+// middleware; scoping is RLS's job (D20) — handlers add no visibility
+// plumbing.
+
+import (
+	"bytes"
+	"net/http"
+
+	"bookhoard/internal/handlers"
+	"bookhoard/templates"
+
+	"github.com/labstack/echo/v5"
+)
+
+func registerPodcastRoutes(cfg *Config) {
+	g := cfg.Echo.Group("/api/podcasts", createJWTMiddleware(cfg))
+
+	// Episode-list renderer (issue #1): htmx actions answer with the
+	// refreshed partial directly — no event indirection to miss. The
+	// view assembly lives here (templates) while the action logic lives
+	// in the handler; this closure is the sanctioned seam.
+	cfg.PodcastHandler.SetEpisodeListRenderer(func(c *echo.Context, feedID string, includePolicySelect, includeOrderSelect bool) error {
+		user, err := getTemplateUserWithTheme(c, cfg)
+		if err != nil {
+			return err
+		}
+		feedUUID, err := parsePodcastUUID(feedID)
+		if err != nil {
+			return err
+		}
+		data, err := buildPodcastDetailData(c, cfg, user, feedUUID)
+		if err != nil {
+			return err
+		}
+		var buf bytes.Buffer
+		render := templates.PodcastEpisodeList
+		if templates.HasActiveDownloads(data.Episodes) {
+			render = templates.PodcastEpisodeListPolling
+		}
+		if err := render(*data).Render(c.Request().Context(), &buf); err != nil {
+			return err
+		}
+		if includePolicySelect {
+			// Issue #3: out-of-band swap of the auto-download select —
+			// the dropdown reflects the flipped policy immediately.
+			if err := templates.PodcastAutoDownloadSelectOOB(data.Feed).Render(c.Request().Context(), &buf); err != nil {
+				return err
+			}
+		}
+		if includeOrderSelect {
+			// §0 item 3: out-of-band swap of the order select — the
+			// "Reverse feed order" button flips to reverse; the
+			// dropdown mirrors the server state immediately.
+			if err := templates.PodcastOrderSelectOOB(data.Feed.ID, data.Order).Render(c.Request().Context(), &buf); err != nil {
+				return err
+			}
+		}
+		return c.HTML(http.StatusOK, buf.String())
+	})
+
+	g.POST("/feeds", cfg.PodcastHandler.AddFeed)
+	g.GET("/feeds", cfg.PodcastHandler.ListFeeds)
+	g.GET("/search", cfg.PodcastHandler.Search)
+
+	// :id-scoped (RLS answers 404 for feeds outside the caller's scope).
+	feeds := g.Group("/feeds/:id")
+	feeds.GET("", cfg.PodcastHandler.GetFeed)
+	feeds.GET("/sequence", cfg.PodcastHandler.GetFeedSequence)
+	feeds.PUT("/order", cfg.PodcastHandler.SetFeedOrder)
+	feeds.PUT("/subscription", cfg.PodcastHandler.UpdateSubscription)
+	feeds.DELETE("/subscription", cfg.PodcastHandler.DeleteSubscription)
+	feeds.POST("/refresh", cfg.PodcastHandler.RefreshFeed)
+	feeds.POST("/episodes/:eid/download", cfg.PodcastHandler.DownloadEpisodes)
+	feeds.DELETE("", cfg.PodcastHandler.DeleteFeed) // admin-only (handler checks role)
+
+	// Admin maintenance (§0 item 8): file deletion frees the disk and
+	// resets rows to pending. Handler checks admin role.
+	feeds.POST("/delete-files", handlers.AdminMiddleware(cfg.PodcastHandler.AdminDeleteFeedFiles))
+	feeds.POST("/episodes/delete-files", handlers.AdminMiddleware(cfg.PodcastHandler.AdminDeleteEpisodeFilesBulk))
+	feeds.POST("/episodes/:eid/delete-file", handlers.AdminMiddleware(cfg.PodcastHandler.AdminDeleteEpisodeFile))
+}

internal/router/podcast_page.go

@@ -0,0 +1,481 @@
+package router
+
+// Podcast SSR pages (plan task 1.9): /podcasts (subscriptions grid +
+// add-feed wizard) and /podcasts/:id (feed detail with the inline
+// download lifecycle). Rendering ownership: ALL content is
+// templ-rendered server data; htmx swaps partials; the sanctioned JS is
+// the modal toggle + event delegation in web/src/podcasts.ts.
+
+import (
+	"bookhoard/internal/database"
+	"bookhoard/internal/handlers"
+	"bookhoard/internal/services"
+	"bookhoard/internal/utils"
+	"bookhoard/templates"
+	"bytes"
+	"encoding/json"
+	"fmt"
+	"net/http"
+	"sort"
+	"strings"
+
+	"github.com/google/uuid"
+	"github.com/jackc/pgx/v5/pgtype"
+	"github.com/labstack/echo/v5"
+)
+
+func registerPodcastPages(cfg *Config) {
+	pages := cfg.Echo.Group("", createJWTMiddleware(cfg))
+
+	pages.GET("/podcasts", func(c *echo.Context) error {
+		user, err := getTemplateUserWithTheme(c, cfg)
+		if err != nil {
+			return renderErrorPage(c, "Error loading user", "user_load_error")
+		}
+		data, err := buildPodcastsPageData(c, cfg)
+		if err != nil {
+			data = &templates.PodcastsPageData{ErrorMsg: "Could not load subscriptions"}
+		}
+		var buf bytes.Buffer
+		if err := templates.PodcastsPage(user, *data, "/podcasts").Render(c.Request().Context(), &buf); err != nil {
+			return err
+		}
+		return c.HTML(http.StatusOK, buf.String())
+	})
+
+	pages.GET("/podcasts/:id", func(c *echo.Context) error {
+		user, err := getTemplateUserWithTheme(c, cfg)
+		if err != nil {
+			return renderErrorPage(c, "Error loading user", "user_load_error")
+		}
+		fid, ferr := parsePodcastUUID(c.Param("id"))
+		if ferr != nil {
+			return c.NoContent(http.StatusNotFound)
+		}
+		data, err := buildPodcastDetailData(c, cfg, user, fid)
+		if err != nil {
+			var buf bytes.Buffer
+			_ = templates.ErrorPage("Feed not found", "not_found").Render(c.Request().Context(), &buf)
+			return c.HTML(http.StatusNotFound, buf.String())
+		}
+		var buf bytes.Buffer
+		if err := templates.PodcastDetailPage(user, *data).Render(c.Request().Context(), &buf); err != nil {
+			return err
+		}
+		return c.HTML(http.StatusOK, buf.String())
+	})
+
+	// Episode list partial — the htmx refresh unit (poll while
+	// downloads are active; podcast:changed events swap it too).
+	pages.GET("/podcasts/:id/episodes", func(c *echo.Context) error {
+		user, err := getTemplateUserWithTheme(c, cfg)
+		if err != nil {
+			return renderErrorPage(c, "Error loading user", "user_load_error")
+		}
+		fid, ferr := parsePodcastUUID(c.Param("id"))
+		if ferr != nil {
+			return c.NoContent(http.StatusNotFound)
+		}
+		data, err := buildPodcastDetailData(c, cfg, user, fid)
+		if err != nil {
+			return c.NoContent(http.StatusNotFound)
+		}
+		var buf bytes.Buffer
+		if templates.HasActiveDownloads(data.Episodes) {
+			err = templates.PodcastEpisodeListPolling(*data).Render(c.Request().Context(), &buf)
+		} else {
+			err = templates.PodcastEpisodeList(*data).Render(c.Request().Context(), &buf)
+		}
+		if err != nil {
+			return err
+		}
+		return c.HTML(http.StatusOK, buf.String())
+	})
+}
+
+// buildPodcastsPageData assembles the /podcasts page: subscription
+// cards, facet values, server-side filters (q/host/tag/language).
+func buildPodcastsPageData(c *echo.Context, cfg *Config) (*templates.PodcastsPageData, error) {
+	user := handlers.MustGetAuthenticatedUser(c)
+	ctx := c.Request().Context()
+
+	libs, err := cfg.Queries.GetPodcastsLibraryForUser(ctx, user.ID)
+	data := &templates.PodcastsPageData{HasLibrary: len(libs) > 0}
+	if err != nil || len(libs) == 0 {
+		return data, err
+	}
+	for _, lib := range libs {
+		data.Libraries = append(data.Libraries, templates.LibraryData{
+			ID:       uuid.UUID(lib.ID.Bytes).String(),
+			Name:     lib.Name,
+			TypeName: "podcasts",
+		})
+	}
+
+	feeds, err := cfg.Queries.ListMyPodcastFeeds(ctx, user.ID)
+	if err != nil {
+		return data, err
+	}
+	// Continue aggregation (plan 2.5): first in-progress episode per
+	// feed (the query is already most-recent-first).
+	continueByFeed := make(map[string]*database.GetPodcastContinueForUserRow)
+	if continueRows, err := cfg.Queries.GetPodcastContinueForUser(ctx, user.ID); err == nil {
+		for i := range continueRows {
+			row := &continueRows[i]
+			key := uuid.UUID(row.SourceFeedID.Bytes).String()
+			if _, ok := continueByFeed[key]; !ok {
+				continueByFeed[key] = row
+			}
+		}
+	}
+	for i := range feeds {
+		row := &feeds[i]
+		total, _ := cfg.Queries.CountPodcastEpisodesForFeed(ctx, row.ID)
+		downloaded, _ := cfg.Queries.CountDownloadedEpisodes(ctx, row.ID)
+		data.Feeds = append(data.Feeds, templates.PodcastFeedCard{
+			ID:                uuid.UUID(row.ID.Bytes).String(),
+			LibraryID:         uuid.UUID(row.LibraryID.Bytes).String(),
+			Title:             row.Title,
+			DisplayName:       row.DisplayName.String,
+			Description:       firstLine(row.Description.String, 220),
+			Host:              row.Host.String,
+			Publisher:         row.Publisher.String,
+			Language:          row.Language.String,
+			Tags:              row.Tags,
+			Explicit:          row.Explicit,
+			ArtworkPath:       row.ArtworkPath.String,
+			FeedType:          row.FeedType.String,
+			Medium:            row.Medium.String,
+			IsComplete:        row.IsComplete,
+			FeedURL:           row.FeedUrl,
+			LastFetched:       pgTimeDisplay(row.LastFetchedAt, utils.PgTextString(user.Timezone)),
+			SubscribedAt:      pgTimeDisplay(row.SubscribedAt, utils.PgTextString(user.Timezone)),
+			EpisodeTotal:      int(total),
+			DownloadedN:       int(downloaded),
+			UndownloadedN:     int(total) - int(downloaded),
+			NewEpisodes:       total-downloaded > 0,
+			AutoDownload:      row.AutoDownload,
+			AutoDownloadCount: int(row.AutoDownloadCount.Int32),
+			Subscribed:        true,
+		})
+		if cont := continueByFeed[uuid.UUID(row.ID.Bytes).String()]; cont != nil {
+			card := &data.Feeds[len(data.Feeds)-1]
+			card.HasContinue = true
+			card.ContinueTitle = cont.Title
+			card.ContinueEpisode = int(cont.FeedPosition.Int32)
+			card.ContinuePercent = cont.Percentage.Float64
+			card.ContinueLastRead = cont.LastReadAt.Time
+		}
+	}
+
+	// Facet values from the unfiltered subscription list (D18).
+	allData := *data
+	data.Facets = facetValuesFromCards(allData.Feeds)
+
+	// Server-side filters.
+	q := strings.ToLower(strings.TrimSpace(c.QueryParam("q")))
+	host := c.QueryParam("host")
+	tag := c.QueryParam("tag")
+	lang := c.QueryParam("language")
+
+	// Episode-level matches (issue #8 / D14): a query also hits episode
+	// titles + descriptions — the feed surfaces with a match count.
+	// SearchMyPodcastEpisodes is subscription-scoped via RLS (the
+	// caller's identity). Only consult it when a query is present.
+	matchesByFeed := make(map[string]int)
+	if q != "" {
+		if eps, err := cfg.Queries.SearchMyPodcastEpisodes(ctx, q); err == nil {
+			for i := range eps {
+				key := uuid.UUID(eps[i].SourceFeedID.Bytes).String()
+				matchesByFeed[key]++
+			}
+		}
+	}
+
+	filtered := data.Feeds[:0:0]
+	for _, f := range data.Feeds {
+		if q != "" {
+			hay := strings.ToLower(f.Title + " " + f.DisplayName + " " + f.Host + " " + f.Publisher + " " + f.Description)
+			matched := matchesByFeed[f.ID]
+			if !strings.Contains(hay, q) && matched == 0 {
+				continue
+			}
+		}
+		if host != "" && f.Host != host {
+			continue
+		}
+		if lang != "" && !strings.EqualFold(f.Language, lang) {
+			continue
+		}
+		if tag != "" {
+			found := false
+			for _, t := range f.Tags {
+				if strings.EqualFold(t, tag) {
+					found = true
+					break
+				}
+			}
+			if !found {
+				continue
+			}
+		}
+		filtered = append(filtered, f)
+	}
+	if q != "" {
+		for i := range filtered {
+			filtered[i].EpisodeMatches = matchesByFeed[filtered[i].ID]
+		}
+	}
+	data.Feeds = filtered
+	data.Query = q
+	data.ContinueFeeds = templates.PodcastContinueFeeds(data.Feeds)
+	return data, nil
+}
+
+// buildPodcastDetailData assembles the feed-detail page + episode list
+// partial. RLS scopes the lookup (non-subscribers get sqlc.ErrNoRows →
+// 404 semantics — no Go visibility plumbing, D20).
+func buildPodcastDetailData(c *echo.Context, cfg *Config, user templates.User, feedID pgtype.UUID) (*templates.PodcastDetailData, error) {
+	actor := handlers.MustGetAuthenticatedUser(c)
+	ctx := c.Request().Context()
+
+	feed, err := cfg.Queries.GetPodcastFeed(ctx, feedID)
+	if err != nil {
+		return nil, err
+	}
+	sub, _ := cfg.Queries.GetPodcastSubscription(ctx, database.GetPodcastSubscriptionParams{
+		UserID: actor.ID, FeedID: feedID,
+	})
+	eps, err := cfg.Queries.ListPodcastEpisodes(ctx, feedID)
+	if err != nil {
+		return nil, err
+	}
+	progressRows, _ := cfg.Queries.GetPodcastEpisodeProgress(ctx, database.GetPodcastEpisodeProgressParams{
+		UserID: actor.ID, SourceFeedID: feedID,
+	})
+	progressByItem := make(map[string]struct {
+		Pct      float64
+		LastRead pgtype.Timestamptz
+	}, len(progressRows))
+	for _, p := range progressRows {
+		progressByItem[uuid.UUID(p.MediaItemID.Bytes).String()] = struct {
+			Pct      float64
+			LastRead pgtype.Timestamptz
+		}{p.Percentage.Float64, p.LastReadAt}
+	}
+
+	total, _ := cfg.Queries.CountPodcastEpisodesForFeed(ctx, feedID)
+	downloaded, _ := cfg.Queries.CountDownloadedEpisodes(ctx, feedID)
+	subscriberCount, _ := cfg.Queries.CountPodcastFeedSubscribers(ctx, feedID)
+
+	card := templates.PodcastFeedCard{
+		ID:                uuid.UUID(feed.ID.Bytes).String(),
+		LibraryID:         uuid.UUID(feed.LibraryID.Bytes).String(),
+		Title:             feed.Title,
+		DisplayName:       sub.DisplayName.String,
+		Description:       feed.Description.String,
+		Host:              feed.Host.String,
+		Publisher:         feed.Publisher.String,
+		Language:          feed.Language.String,
+		Tags:              feed.Tags,
+		Explicit:          feed.Explicit,
+		ArtworkPath:       feed.ArtworkPath.String,
+		FeedType:          feed.FeedType.String,
+		Medium:            feed.Medium.String,
+		IsComplete:        feed.IsComplete,
+		FeedURL:           feed.FeedUrl,
+		LastFetched:       pgTimeDisplay(feed.LastFetchedAt, user.Timezone),
+		EpisodeTotal:      int(total),
+		DownloadedN:       int(downloaded),
+		UndownloadedN:     int(total) - int(downloaded),
+		NewEpisodes:       total-downloaded > 0,
+		AutoDownload:      sub.AutoDownload,
+		AutoDownloadCount: int(sub.AutoDownloadCount.Int32),
+		Subscribed:        sub.ID.Valid,
+	}
+
+	rows := make([]templates.PodcastEpisodeRow, 0, len(eps))
+	gaps := 0
+	seenDownloadedFrom := false // walk reversed: gap = undownloaded with a downloaded LATER in order
+	for i := len(eps) - 1; i >= 0; i-- {
+		ep := &eps[i]
+		downloadedEp := ep.DownloadState == "downloaded"
+		if !downloadedEp && seenDownloadedFrom {
+			gaps++
+		}
+		if downloadedEp {
+			seenDownloadedFrom = true
+		}
+	}
+	// Forward pass for the actual rows (native order — the list keeps
+	// feed order; the ORDER selector drives the sequence endpoint, and
+	// the order control re-renders this list via the override applied
+	// below).
+	order := podcastOrderForFeed(cfg, c, feedID.String())
+	for i := range eps {
+		ep := &eps[i]
+		rowID := uuid.UUID(ep.ID.Bytes).String()
+		p := progressByItem[rowID]
+		dateDisplay := ""
+		dateSort := ""
+		if ep.DatePublished.Valid {
+			dateSort = ep.DatePublished.Time.Format("2006-01-02")
+			dateDisplay = ep.DatePublished.Time.In(utils.UserLocation(user.Timezone)).Format("Jan 2, 2006")
+		}
+		rows = append(rows, templates.PodcastEpisodeRow{
+			ID:            uuid.UUID(ep.ID.Bytes).String(),
+			FeedID:        uuid.UUID(ep.SourceFeedID.Bytes).String(),
+			Title:         ep.Title,
+			Description:   firstLine(ep.Description.String, 200),
+			Date:          dateDisplay,
+			SortDate:      dateSort,
+			Position:      int(ep.FeedPosition.Int32),
+			Duration:      templates.FormatEpisodeDuration(int(ep.DeclaredDurationSeconds.Int32)),
+			Size:          templates.FormatEpisodeSize(ep.FileSize.Int64),
+			PlayURL:       "/readers/" + uuid.UUID(ep.ID.Bytes).String(),
+			ListenPct:     p.Pct,
+			Downloaded:    ep.DownloadState == "downloaded",
+			DownloadState: ep.DownloadState,
+			DownloadError: ep.DownloadError.String,
+		})
+	}
+	// Order override reshapes the LIST too (the sequence endpoint is the
+	// strict downloaded-only view; the list shows everything in the
+	// chosen order so pending episodes stay visible). Consolidated sort
+	// (§0 item 5a) — SortDate is the lexical pub-date key so the
+	// tiebreak is chronological even in display-tz date strings.
+	services.OrderEpisodes(rows, order,
+		func(r templates.PodcastEpisodeRow) int32 { return int32(r.Position) },
+		func(r templates.PodcastEpisodeRow) string { return r.SortDate },
+	)
+
+	detail := &templates.PodcastDetailData{
+		Feed:            card,
+		Episodes:        rows,
+		GapCount:        gaps,
+		Order:           order,
+		IsAdmin:         actor.Role == "admin",
+		IsSerial:        feed.FeedType.String == "serial" || feed.Medium.String == "audiobook",
+		SubscriberCount: int(subscriberCount),
+	}
+
+	// Live byte progress for in-flight downloads (issue #1): read the
+	// worker's snapshot; rows not actively downloading render states.
+	if cfg.PodcastWorker != nil {
+		for i := range detail.Episodes {
+			row := &detail.Episodes[i]
+			if row.DownloadState != "downloading" {
+				continue
+			}
+			if recv, tot, ok := cfg.PodcastWorker.EpisodeProgress(pgtype.UUID{Bytes: uuid.MustParse(row.ID), Valid: true}); ok {
+				row.ProgressKnown = true
+				row.ProgressReceived = recv
+				row.ProgressTotal = tot
+				if tot > 0 {
+					row.ProgressPct = float64(recv) / float64(tot) * 100
+				}
+			}
+		}
+	}
+	return detail, nil
+}
+
+// podcastOrderForFeed reads the caller's order override (D5).
+//
+// The UNSET default is "reverse" (2026-11-01 user request): the platform
+// is audiobook-first and feed document order runs newest-first, so
+// native starts a book at its LAST chapter. This is the same per-feed
+// setting the Order dropdown shows — page list, player chain, and
+// download queue all resolve through it. Stored overrides are untouched:
+// anyone who picked an order keeps exactly what they picked.
+func podcastOrderForFeed(cfg *Config, c *echo.Context, feedID string) string {
+	user := handlers.MustGetAuthenticatedUser(c)
+	row, err := cfg.Queries.GetUserSettingByKey(c.Request().Context(), database.GetUserSettingByKeyParams{
+		UserID:     user.ID,
+		SettingKey: "podcast_order:" + feedID,
+	})
+	if err != nil {
+		return "reverse"
+	}
+	var v struct {
+		Order string `json:"order"`
+	}
+	if err := json.Unmarshal(row, &v); err != nil {
+		return "reverse"
+	}
+	if !services.ValidPodcastEpisodeOrder(v.Order) {
+		return "reverse"
+	}
+	return v.Order
+}
+
+// facetValuesFromCards derives the filter bar values.
+// facetValuesFromCards derives the filter-bar values with feed counts,
+// sorted by frequency DESC then alphabetically — the top of each group
+// is what a user with hundreds of feeds actually filters by.
+func facetValuesFromCards(feeds []templates.PodcastFeedCard) map[string][]templates.FacetValue {
+	hosts, tags, langs := map[string]int{}, map[string]int{}, map[string]int{}
+	for _, f := range feeds {
+		if f.Host != "" {
+			hosts[f.Host]++
+		}
+		for _, t := range f.Tags {
+			tags[t]++
+		}
+		if f.Language != "" {
+			langs[f.Language]++
+		}
+	}
+	return map[string][]templates.FacetValue{
+		"hosts":    sortedFacet(hosts),
+		"tags":     sortedFacet(tags),
+		"language": sortedFacet(langs),
+	}
+}
+
+// sortedFacet: count DESC, then value ASC (stable, scannable).
+func sortedFacet(set map[string]int) []templates.FacetValue {
+	out := make([]templates.FacetValue, 0, len(set))
+	for k, n := range set {
+		out = append(out, templates.FacetValue{Value: k, Count: n})
+	}
+	sort.Slice(out, func(i, j int) bool {
+		if out[i].Count != out[j].Count {
+			return out[i].Count > out[j].Count
+		}
+		return out[i].Value < out[j].Value
+	})
+	return out
+}
+
+func firstLine(s string, max int) string {
+	s = strings.TrimSpace(s)
+	if i := strings.IndexAny(s, "\n\r"); i > 0 {
+		s = s[:i]
+	}
+	if len(s) > max {
+		s = s[:max] + "…"
+	}
+	return s
+}
+
+func pgTimeDisplay(t pgtype.Timestamptz, tz string) string {
+	if !t.Valid {
+		return ""
+	}
+	return utils.FormatTimeInTimezone(t.Time, tz)
+}
+
+func itoaPad(n int) string {
+	return fmt.Sprintf("%06d", n)
+}
+
+// parsePodcastUUID parses an id path param (handlers.parseUUID is
+// package-private).
+func parsePodcastUUID(s string) (pgtype.UUID, error) {
+	id, err := uuid.Parse(strings.TrimSpace(s))
+	if err != nil {
+		return pgtype.UUID{}, err
+	}
+	return pgtype.UUID{Bytes: id, Valid: true}, nil
+}

internal/router/reader.go

@@ -1,13 +1,16 @@
 package router
 
 import (
+	"bookhoard/internal/database"
 	"bookhoard/internal/handlers"
 	"bookhoard/internal/services"
 	"bookhoard/internal/sync"
 	"bookhoard/internal/utils"
 	"bookhoard/templates"
 	"bytes"
+	"encoding/json"
 	"net/http"
+	"strings"
 
 	"github.com/google/uuid"
 	"github.com/jackc/pgx/v5/pgtype"
@@ -102,6 +105,18 @@ func registerReaderRoutes(cfg *Config) {
 			TreatAsComic:     handlers.ShouldTreatAsComic(libraryWithType.TypeName, mediaItem.FormatGroup),
 			KeepScreenAwake:  user.KeepScreenAwake,
 		}
+		// Podcast chaining (plan D7): episodes render the audio reader
+		// with sequence context — the neighboring DOWNLOADED episodes in
+		// the caller's effective order (override applied). RLS already
+		// scoped the episode read; the subscription predicate covers the
+		// neighbors the same way. ?single=1 escapes chaining (the "play
+		// single episode" affordance).
+		if mediaItem.Source == "podcast_episode" && mediaItem.SourceFeedID.Valid && c.QueryParam("single") != "1" {
+			userPG := uuidToPGType(userUUID)
+			if seq := buildPodcastSequenceContext(c, cfg, &mediaItem, userPG, libraryWithType.TypeName); seq != nil {
+				metadata.PodcastSequence = seq
+			}
+		}
 		// Render template
 		var buf bytes.Buffer
 		// Audiobooks get the audio player page — the foliate document
@@ -144,3 +159,276 @@ func registerReaderRoutes(cfg *Config) {
 	reader.GET("/settings", cfg.ReaderHandler.GetSettings)
 	reader.PUT("/settings", cfg.ReaderHandler.UpdateSettings)
 }
+
+// buildPodcastSequenceContext computes the episode chain payload (plan
+// D7): effective order — the D5 override applied, feed-native fallback —
+// DOWNLOADED episodes only (gaps never enter the queue), the current
+// episode's index, and prev/next reader URLs. RLS scopes the episode
+// reads; non-subscribers never reach this path in the first place.
+func buildPodcastSequenceContext(c *echo.Context, cfg *Config, item *database.MediaItems, userID pgtype.UUID, libraryType string) *templates.PodcastSequenceContext {
+	ctx := c.Request().Context()
+	// Queue override (plan item 2.3): ?queue=collection:<id> plays the
+	// episode inside an Up Next collection chain. Absent/malformed/
+	// not-owned ids fall back to the feed queue — the reader page must
+	// never hard-fail on a stale link.
+	if q := c.QueryParam("queue"); strings.HasPrefix(q, "collection:") {
+		if collID, err := uuid.Parse(strings.TrimPrefix(q, "collection:")); err == nil {
+			if seq := buildCollectionSequenceContext(c, cfg, item, userID, collID); seq != nil {
+				return seq
+			}
+		}
+	}
+	eps, err := cfg.Queries.ListPodcastEpisodes(ctx, item.SourceFeedID)
+	if err != nil || len(eps) == 0 {
+		return nil
+	}
+	feed, err := cfg.Queries.GetPodcastFeed(ctx, item.SourceFeedID)
+	if err != nil {
+		return nil
+	}
+
+	// Effective order override (D5): reader_settings KV. UNSET defaults
+	// to "reverse" (2026-11-01: audiobook-first — native document order
+	// starts a book at its last chapter); the same default the podcast
+	// page's Order dropdown shows. Stored overrides are untouched.
+	order := "reverse"
+	if row, err := cfg.Queries.GetUserSettingByKey(ctx, database.GetUserSettingByKeyParams{
+		UserID:     userID,
+		SettingKey: "podcast_order:" + uuid.UUID(item.SourceFeedID.Bytes).String(),
+	}); err == nil {
+		var v struct {
+			Order string `json:"order"`
+		}
+		if json.Unmarshal(row, &v) == nil && services.ValidPodcastEpisodeOrder(v.Order) {
+			order = v.Order
+		}
+	}
+
+	type seqEp struct {
+		id    string
+		title string
+		date  string
+		pos   int32
+	}
+	ordered := make([]seqEp, 0, len(eps))
+	currentID := uuid.UUID(item.ID.Bytes).String()
+	for i := range eps {
+		ep := eps[i]
+		if ep.DownloadState != "downloaded" {
+			continue // un-downloaded episodes are skipped, not tracked (invariant 6)
+		}
+		date := ""
+		if ep.DatePublished.Valid {
+			date = ep.DatePublished.Time.Format("2006-01-02")
+		}
+		ordered = append(ordered, seqEp{
+			id:    uuid.UUID(ep.ID.Bytes).String(),
+			title: ep.Title,
+			date:  date,
+			pos:   ep.FeedPosition.Int32,
+		})
+	}
+	// Effective listening order (order consolidation §0 item 5a).
+	services.OrderEpisodes(ordered, order,
+		func(e seqEp) int32 { return e.pos },
+		func(e seqEp) string { return e.date },
+	)
+
+	idx := -1
+	for i, ep := range ordered {
+		if ep.id == currentID {
+			idx = i
+			break
+		}
+	}
+	if idx < 0 {
+		return nil // current episode not downloaded (or not in sequence)
+	}
+
+	seq := &templates.PodcastSequenceContext{
+		FeedURL:   "/podcasts/" + uuid.UUID(item.SourceFeedID.Bytes).String(),
+		FeedTitle: feed.Title,
+		// R2: this producer is the feed queue. The player's order
+		// control renders only for this source.
+		Source:       "feed",
+		FeedID:       uuid.UUID(item.SourceFeedID.Bytes).String(),
+		Order:        order,
+		EpisodeIndex: idx + 1,
+		IndexTotal:   len(ordered),
+	}
+	if idx > 0 {
+		seq.PrevURL = "/readers/" + ordered[idx-1].id
+	}
+	if idx < len(ordered)-1 {
+		seq.NextURL = "/readers/" + ordered[idx+1].id
+		seq.NextTitle = ordered[idx+1].title
+	}
+	return seq
+}
+
+// buildCollectionSequenceContext is the COLLECTION queue producer (plan
+// item 2.3, §10 R2: same context shape, different owner). The queue is
+// the collection's ordered downloaded episodes — the collection's
+// order IS content (R1), so no feed-order override applies here and
+// the player's order control stays hidden (source=collection).
+// Ownership gates the chain: collections are user-scoped app state
+// (no RLS on the table), and a foreign or deleted collection id falls
+// back to the feed queue. Prev/next URLs carry ?queue=collection:<id>
+// so the chain survives hops; backToDetail lands on the collection.
+func buildCollectionSequenceContext(c *echo.Context, cfg *Config, item *database.MediaItems, userID pgtype.UUID, collID uuid.UUID) *templates.PodcastSequenceContext {
+	ctx := c.Request().Context()
+	owns, err := cfg.CollectionHandler.CollectionService().UserOwnsCollection(ctx, collID, uuid.UUID(userID.Bytes))
+	if err != nil || !owns {
+		return nil
+	}
+	coll, err := cfg.Queries.GetCollection(ctx, pgtype.UUID{Bytes: collID, Valid: true})
+	if err != nil {
+		return nil
+	}
+	// Smart collections compute their membership live (R1: computed
+	// order, never written back) — evaluate the rules for THIS viewer
+	// and chain the downloaded episode matches in native feed order
+	// (R3: every listening order routes through OrderEpisodes).
+	if services.IsCollectionSmart(coll) {
+		return buildSmartCollectionSequenceContext(c, cfg, item, userID, coll)
+	}
+	// Downloaded episodes in collection order (SQL orders; RLS scopes
+	// the join — unsubscribed feeds drop out and the chain skips them).
+	eps, err := cfg.Queries.GetCollectionEpisodesOrdered(ctx, pgtype.UUID{Bytes: collID, Valid: true})
+	if err != nil || len(eps) == 0 {
+		return nil
+	}
+	// The curated listening order applies to static episode playlists
+	// too (§10 R2): native/"" keeps the manual position order (the SQL
+	// order above); reverse flips it; oldest/newest re-key on
+	// publication date with feed position as the tiebreak.
+	switch order := handlers.CollectionChainOrder(coll.ViewSettings); order {
+	case "reverse":
+		for i, j := 0, len(eps)-1; i < j; i, j = i+1, j-1 {
+			eps[i], eps[j] = eps[j], eps[i]
+		}
+	case "oldest", "newest":
+		services.OrderEpisodes(eps, order,
+			func(e database.GetCollectionEpisodesOrderedRow) int32 { return e.FeedPosition.Int32 },
+			func(e database.GetCollectionEpisodesOrderedRow) string {
+				if e.DatePublished.Valid {
+					return e.DatePublished.Time.Format("2006-01-02")
+				}
+				return ""
+			})
+	}
+
+	currentID := uuid.UUID(item.ID.Bytes).String()
+	idx := -1
+	for i, ep := range eps {
+		if uuid.UUID(ep.ID.Bytes).String() == currentID {
+			idx = i
+			break
+		}
+	}
+	if idx < 0 {
+		return nil // current episode not downloaded (or not in sequence)
+	}
+
+	queueParam := "?queue=collection:" + collID.String()
+	seq := &templates.PodcastSequenceContext{
+		FeedURL:      "/collections/" + collID.String(),
+		FeedTitle:    coll.Name,
+		Source:       "collection",
+		EpisodeIndex: idx + 1,
+		IndexTotal:   len(eps),
+	}
+	if idx > 0 {
+		seq.PrevURL = "/readers/" + uuid.UUID(eps[idx-1].ID.Bytes).String() + queueParam
+	}
+	if idx < len(eps)-1 {
+		seq.NextURL = "/readers/" + uuid.UUID(eps[idx+1].ID.Bytes).String() + queueParam
+		seq.NextTitle = eps[idx+1].Title
+	}
+	return seq
+}
+
+// buildSmartCollectionSequenceContext chains a SMART collection's live
+// membership (plan item 2.5 acceptance: "Daily Fiction" → Play =
+// today's episodes back-to-back). The rules evaluate for the requesting
+// user (mode from the collection; family lens via the visible set);
+// episode matches that are downloaded chain in native feed order.
+// Books and un-downloaded matches never enter the queue (invariant 6).
+func buildSmartCollectionSequenceContext(c *echo.Context, cfg *Config, item *database.MediaItems, userID pgtype.UUID, coll database.Collections) *templates.PodcastSequenceContext {
+	ctx := c.Request().Context()
+	colService := cfg.CollectionHandler.CollectionService()
+	visibleIDs, err := cfg.CollectionHandler.UserVisibleLibraryIDs(c)
+	if err != nil {
+		visibleIDs = nil
+	}
+	matches, err := colService.EvaluateSmartMembership(ctx, coll, services.SmartEvalOptions{
+		VisibleIDs:   visibleIDs,
+		ViewerUserID: userID,
+	})
+	if err != nil {
+		return nil
+	}
+
+	type seqEp struct {
+		id    string
+		title string
+		date  string
+		pos   int32
+	}
+	ordered := make([]seqEp, 0, len(matches))
+	currentID := uuid.UUID(item.ID.Bytes).String()
+	for _, m := range matches {
+		if m.Source != "podcast_episode" || m.DownloadState != "downloaded" {
+			continue
+		}
+		date := ""
+		if m.DatePublished.Valid {
+			date = m.DatePublished.Time.Format("2006-01-02")
+		}
+		ordered = append(ordered, seqEp{
+			id:    uuid.UUID(m.ID.Bytes).String(),
+			title: m.Title,
+			date:  date,
+			pos:   m.FeedPosition.Int32,
+		})
+	}
+	if len(ordered) == 0 {
+		return nil
+	}
+	// Plan §10 R2: the collection's CURATED listening order (stored in
+	// view_settings.chain_order, set from the detail header) — not a
+	// player-side knob. "" / native / unknown all mean feed order.
+	services.OrderEpisodes(ordered, handlers.CollectionChainOrder(coll.ViewSettings),
+		func(e seqEp) int32 { return e.pos },
+		func(e seqEp) string { return e.date },
+	)
+
+	idx := -1
+	for i, ep := range ordered {
+		if ep.id == currentID {
+			idx = i
+			break
+		}
+	}
+	if idx < 0 {
+		return nil
+	}
+
+	collID := uuid.UUID(coll.ID.Bytes).String()
+	queueParam := "?queue=collection:" + collID
+	seq := &templates.PodcastSequenceContext{
+		FeedURL:      "/collections/" + collID,
+		FeedTitle:    coll.Name,
+		Source:       "collection",
+		EpisodeIndex: idx + 1,
+		IndexTotal:   len(ordered),
+	}
+	if idx > 0 {
+		seq.PrevURL = "/readers/" + ordered[idx-1].id + queueParam
+	}
+	if idx < len(ordered)-1 {
+		seq.NextURL = "/readers/" + ordered[idx+1].id + queueParam
+		seq.NextTitle = ordered[idx+1].title
+	}
+	return seq
+}

internal/router/router.go

@@ -80,6 +80,9 @@ type Config struct {
 	SidecarHandler          *handlers.SidecarHandler
 	ReaderHandler           *handlers.ReaderHandler
 	LibraryService          *services.LibraryService
+	PodcastHandler          *handlers.PodcastHandler
+	PodcastService          *services.PodcastService
+	PodcastWorker           *services.PodcastDownloadWorker
 }
 
 // getBaseURL returns the configured base URL from the database, falling back to
@@ -344,12 +347,14 @@ func RegisterRoutes(cfg *Config) *handlers.Handler {
 	registerConflictRoutes(cfg)
 	registerAnalyticsRoutes(cfg)
 	registerQueueRoutes(cfg)
+	registerPodcastRoutes(cfg)
 	registerJobRoutes(cfg)
 	registerFiltersRoutes(cfg)
 	registerOPDSRoutes(cfg)
 	registerReaderRoutes(cfg)
 	registerWebSocketRoutes(cfg)
 	registerFrontendRoutes(cfg)
+	registerPodcastPages(cfg)
 	registerDocumentationRoutes(cfg)
 
 	// Static bundles are served unhashed (reader.js, index.js), so the
@@ -394,6 +399,14 @@ func RegisterRoutes(cfg *Config) *handlers.Handler {
 	// Start background tasks (queue processor and connection cleanup)
 	scannerHandler.StartBackgroundTasks()
 
+	// Podcast subsystem background loops: sequential download worker +
+	// feed poller (plan §7). The poller re-reads its interval setting
+	// every tick; both exit with the server context.
+	if cfg.PodcastWorker != nil && cfg.PodcastService != nil {
+		cfg.PodcastWorker.Start(context.Background())
+		services.NewPodcastPoller(cfg.Queries, cfg.PodcastService, cfg.PodcastWorker, cfg.Settings).Start(context.Background())
+	}
+
 	// Start watch mode for all libraries (after 2 second delay for DB)
 	go func() {
 		time.Sleep(2 * time.Second)

internal/router/search.go

@@ -16,6 +16,10 @@ func registerSearchRoutes(cfg *Config) {
 	// JWT middleware for protected routes
 	jwtMiddleware := createJWTMiddleware(cfg)
 	protected := e.Group("/api", jwtMiddleware)
+	// Filter label resolution (2026-11-01): names id-kind filter values
+	// from their source tables for trees the server didn't render (the
+	// cold saved-filter apply). Visibility/RLS ride the request identity.
+	protected.GET("/filter/labels", cfg.MediaHandler.ResolveFilterLabels)
 	// Search endpoint - returns JSON or HTML based on HX-Request header
 	protected.GET("/media-items/search", func(c *echo.Context) error {
 		// Check if HTMX request (expects HTML)
@@ -41,6 +45,24 @@ func handleSearchHTML(c *echo.Context, cfg *Config) error {
 	if !ok {
 		return c.String(http.StatusUnauthorized, "Unauthorized")
 	}
+
+	// Episode picker scope (plan item 2.3): ?scope=episodes&show_checkbox=true
+	// swaps the book cards for YOUR subscriptions' episodes — visibility
+	// is the RLS policy's job (D20), the Go side adds nothing. The books
+	// branch keeps its podcasts exclusion; this is the only door in.
+	if c.QueryParam("scope") == "episodes" && c.QueryParam("show_checkbox") == "true" {
+		eps, err := cfg.Queries.SearchEpisodesForPicker(c.Request().Context(), c.QueryParam("q"))
+		if err != nil {
+			log.Printf("Episode picker search error: %v", err)
+			return c.HTML(http.StatusInternalServerError, `<div style="color: red;">Search error</div>`)
+		}
+		var buf bytes.Buffer
+		if err := templates.EpisodePickerGrid(eps).Render(c.Request().Context(), &buf); err != nil {
+			return err
+		}
+		return c.HTML(http.StatusOK, buf.String())
+	}
+
 	// Extract filters via the SHARED parser (handlers.ParseMediaSearchParams)
 	// — the HTMX grid partial and the JSON endpoint can never disagree on
 	// which facet params exist or how they parse. (This branch previously

internal/services/collection_service.go

@@ -6,7 +6,9 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
+	"strconv"
 	"strings"
+	"time"
 
 	"github.com/google/uuid"
 	"github.com/jackc/pgx/v5"
@@ -30,24 +32,6 @@ type Rule struct {
 	Enabled *bool `json:"enabled,omitempty"`
 }
 
-// Known rule fields and operators — the rules builder UI and the
-// persisted JSONB documents must agree with this vocabulary.
-var ruleFields = map[string]bool{
-	"genre": true, "series": true, "author": true, "language": true,
-	"publisher": true, "copyright_year": true, "tags": true,
-}
-
-var ruleOperators = map[string]bool{
-	"equals": true, "not_equals": true, "contains": true, "not_contains": true,
-	"starts_with": true, "ends_with": true, "greater_than": true, "less_than": true,
-}
-
-// RuleValidationError reports a rule payload the evaluator could never
-// honor (unknown field/operator, empty value).
-type RuleValidationError struct{ Msg string }
-
-func (e *RuleValidationError) Error() string { return e.Msg }
-
 // RuleEvaluation represents the result of evaluating a rule
 type RuleEvaluation struct {
 	RuleID     string  `json:"rule_id"`
@@ -58,27 +42,28 @@ type RuleEvaluation struct {
 // CollectionService handles collection management
 type CollectionService struct {
 	db *database.Queries
+	// search powers the unified-engine path (item 3.2): tree-driven
+	// smart collections evaluate via ComposedLiveMedia. Optional —
+	// nil disables tree evaluation (legacy-only services, which keeps
+	// every existing constructor call site valid).
+	search *SearchService
 }
 
-// NewCollectionService creates a new collection service
+// NewCollectionService creates a new collection service. search is
-func NewCollectionService(db *database.Queries) *CollectionService {
+// optional (variadic for constructor compatibility) — required for
-	return &CollectionService{
+// TREE-driven smart collections (item 3.2), ignored otherwise.
-		db: db,
+func NewCollectionService(db *database.Queries, search ...*SearchService) *CollectionService {
+	svc := &CollectionService{db: db}
+	if len(search) > 0 {
+		svc.search = search[0]
 	}
+	return svc
 }
 
-// CreateCollection creates a new collection
+// CreateCollection creates a new collection. filterTree (item 3.2), when
-func (s *CollectionService) CreateCollection(ctx context.Context, userID uuid.UUID, name, description, color, icon string, autoAssignRules []Rule, viewSettings map[string]any) (database.Collections, error) {
+// present, makes it smart — the qf engine is the only membership source
-	// Convert rules to JSONB ([]byte)
+// (the legacy rules-array storage retired 2026-10-31).
-	var rulesJSON []byte
+func (s *CollectionService) CreateCollection(ctx context.Context, userID uuid.UUID, name, description, color, icon string, viewSettings map[string]any, showOnDashboard bool, filterTree []byte) (database.Collections, error) {
-	if len(autoAssignRules) > 0 {
-		var err error
-		rulesJSON, err = json.Marshal(autoAssignRules)
-		if err != nil {
-			return database.Collections{}, fmt.Errorf("failed to marshal auto-assign rules: %v", err)
-		}
-	}
-
 	// Convert view settings to JSONB ([]byte)
 	var settingsJSON []byte
 	if len(viewSettings) > 0 {
@@ -95,8 +80,9 @@ func (s *CollectionService) CreateCollection(ctx context.Context, userID uuid.UU
 		Description:     pgtype.Text{String: description, Valid: description != ""},
 		Color:           pgtype.Text{String: color, Valid: color != ""},
 		Icon:            pgtype.Text{String: icon, Valid: icon != ""},
-		AutoAssignRules: rulesJSON,
 		ViewSettings:    settingsJSON,
+		FilterTree:      filterTree,
+		ShowOnDashboard: pgtype.Bool{Bool: showOnDashboard, Valid: true},
 	})
 
 	if err != nil {
@@ -132,14 +118,17 @@ func (s *CollectionService) GetUserCollections(ctx context.Context, userID uuid.
 	return collections, nil
 }
 
-// UpdateCollection updates a collection
+// UpdateCollection rewrites the collection row. filterTree semantics:
-func (s *CollectionService) UpdateCollection(ctx context.Context, collectionID uuid.UUID, name, description, color, icon string, autoAssignRules []Rule, viewSettings map[string]any) (database.Collections, error) {
+// nil = leave the stored tree untouched (forms don't send trees).
+func (s *CollectionService) UpdateCollection(ctx context.Context, collectionID uuid.UUID, name, description, color, icon string, viewSettings map[string]any, filterTree []byte) (database.Collections, error) {
 	// A2c §3.3: the edit form carries only the toggles it renders — merge
 	// them OVER the stored view_settings so provenance keys
 	// (imported_from_series) and anything else on the row survive.
-	if len(viewSettings) > 0 {
+	if existing, err := s.db.GetCollection(ctx, pgtype.UUID{Bytes: collectionID, Valid: true}); err == nil {
-		existing, err := s.db.GetCollection(ctx, pgtype.UUID{Bytes: collectionID, Valid: true})
+		if filterTree == nil {
-		if err == nil && len(existing.ViewSettings) > 0 {
+			filterTree = existing.FilterTree // nil = untouched (forms don't send trees)
+		}
+		if len(viewSettings) > 0 && len(existing.ViewSettings) > 0 {
 			var stored map[string]any
 			if json.Unmarshal(existing.ViewSettings, &stored) == nil {
 				merged := stored
@@ -149,18 +138,18 @@ func (s *CollectionService) UpdateCollection(ctx context.Context, collectionID u
 				viewSettings = merged
 			}
 		}
+		// Same convention as filterTree: a NIL view_settings payload
+		// means untouched — keep the stored doc. Without this, an edit
+		// that carries no settings wrote NULL over the row and silently
+		// dropped every stored key (chain_order, imported provenance);
+		// the 2026-10-31 chain-order test pinned exactly this clobber.
+		if viewSettings == nil && len(existing.ViewSettings) > 0 {
+			var stored map[string]any
+			if json.Unmarshal(existing.ViewSettings, &stored) == nil && len(stored) > 0 {
+				viewSettings = stored
 			}
-
-	// Convert rules to JSONB ([]byte)
-	var rulesJSON []byte
-	if len(autoAssignRules) > 0 {
-		var err error
-		rulesJSON, err = json.Marshal(autoAssignRules)
-		if err != nil {
-			return database.Collections{}, fmt.Errorf("failed to marshal auto-assign rules: %v", err)
 		}
 	}
-
 	// Convert view settings to JSONB ([]byte)
 	var settingsJSON []byte
 	if len(viewSettings) > 0 {
@@ -177,12 +166,12 @@ func (s *CollectionService) UpdateCollection(ctx context.Context, collectionID u
 		Description:  pgtype.Text{String: description, Valid: description != ""},
 		Color:        pgtype.Text{String: color, Valid: color != ""},
 		Icon:         pgtype.Text{String: icon, Valid: icon != ""},
-		AutoAssignRules: rulesJSON,
 		ViewSettings: settingsJSON,
+		FilterTree:   filterTree,
 	})
 
 	if err != nil {
-		return database.Collections{}, fmt.Errorf("failed to update collection: %v", err)
+		return database.Collections{}, fmt.Errorf("failed to update collection: %w", err)
 	}
 
 	return collection, nil
@@ -260,162 +249,7 @@ func (s *CollectionService) GetBookCollections(ctx context.Context, mediaItemID
 	}
 
 	return collections, nil
-}
+} // EnabledRules filters a ruleset down to the rules that participate in// validateRule checks a rule payload the evaluator will have to honor.// AddRule appends a rule to the collection's ruleset, generating an id// SetRuleEnabled flips a single rule's enabled flag in place.// RemoveRule deletes a single rule from the ruleset. Removing the LAST// SmartEvalOptions scopes a smart collection's live evaluation.
-
-// ParseRules decodes a collection's auto_assign_rules JSONB into typed
-// rules. "null", empty, and malformed docs mean "no rules".
-func (s *CollectionService) ParseRules(raw []byte) []Rule {
-	if len(raw) == 0 {
-		return nil
-	}
-	var rules []Rule
-	if err := json.Unmarshal(raw, &rules); err != nil {
-		return nil
-	}
-	return rules
-}
-
-// IsSmartCollection derives the collection type: a non-empty ruleset
-// makes it SMART (live-matched), an empty/absent ruleset makes it STATIC
-// (stored collection_items links).
-func (s *CollectionService) IsSmartCollection(raw []byte) bool {
-	return len(s.ParseRules(raw)) > 0
-}
-
-// EnabledRules filters a ruleset down to the rules that participate in
-// live evaluation (absent enabled = enabled for backward compatibility).
-func (s *CollectionService) EnabledRules(rules []Rule) []Rule {
-	enabled := make([]Rule, 0, len(rules))
-	for _, rule := range rules {
-		if rule.Enabled == nil || *rule.Enabled {
-			enabled = append(enabled, rule)
-		}
-	}
-	return enabled
-}
-
-// validateRule checks a rule payload the evaluator will have to honor.
-func (s *CollectionService) validateRule(rule Rule) error {
-	if !ruleFields[rule.Field] {
-		return &RuleValidationError{Msg: "unknown rule field: " + rule.Field}
-	}
-	if !ruleOperators[rule.Operator] {
-		return &RuleValidationError{Msg: "unknown rule operator: " + rule.Operator}
-	}
-	if strings.TrimSpace(rule.Value) == "" {
-		return &RuleValidationError{Msg: "rule value is required"}
-	}
-	if rule.Priority < 0 || rule.Priority > 10 {
-		return &RuleValidationError{Msg: "priority must be between 0 and 10"}
-	}
-	return nil
-}
-
-// AddRule appends a rule to the collection's ruleset, generating an id
-// when the caller did not supply one. Adding the first rule CONVERTS a
-// static collection to smart (type is derived from rules presence).
-func (s *CollectionService) AddRule(ctx context.Context, collectionID uuid.UUID, rule Rule) (database.Collections, Rule, error) {
-	if err := s.validateRule(rule); err != nil {
-		return database.Collections{}, Rule{}, err
-	}
-
-	collection, err := s.GetCollection(ctx, collectionID)
-	if err != nil {
-		return database.Collections{}, Rule{}, err
-	}
-
-	rules := s.ParseRules(collection.AutoAssignRules)
-	if rule.ID == "" {
-		rule.ID = uuid.NewString()
-	}
-	rules = append(rules, rule)
-
-	rulesJSON, marshalErr := json.Marshal(rules)
-	if marshalErr != nil {
-		return database.Collections{}, Rule{}, fmt.Errorf("failed to marshal rules: %w", marshalErr)
-	}
-
-	updated, err := s.db.UpdateCollectionRulesJSON(ctx, database.UpdateCollectionRulesJSONParams{
-		ID:              pgtype.UUID{Bytes: collectionID, Valid: true},
-		AutoAssignRules: rulesJSON,
-	})
-	if err != nil {
-		return database.Collections{}, Rule{}, fmt.Errorf("failed to save rules: %w", err)
-	}
-	return updated, rule, nil
-}
-
-// SetRuleEnabled flips a single rule's enabled flag in place.
-func (s *CollectionService) SetRuleEnabled(ctx context.Context, collectionID uuid.UUID, ruleID string, enabled bool) (database.Collections, error) {
-	collection, err := s.GetCollection(ctx, collectionID)
-	if err != nil {
-		return database.Collections{}, err
-	}
-	rules := s.ParseRules(collection.AutoAssignRules)
-	found := false
-	for i := range rules {
-		if rules[i].ID == ruleID {
-			rules[i].Enabled = &enabled
-			found = true
-			break
-		}
-	}
-	if !found {
-		return database.Collections{}, fmt.Errorf("rule not found: %s", ruleID)
-	}
-
-	rulesJSON, marshalErr := json.Marshal(rules)
-	if marshalErr != nil {
-		return database.Collections{}, fmt.Errorf("failed to marshal rules: %w", marshalErr)
-	}
-
-	updated, err := s.db.UpdateCollectionRulesJSON(ctx, database.UpdateCollectionRulesJSONParams{
-		ID:              pgtype.UUID{Bytes: collectionID, Valid: true},
-		AutoAssignRules: rulesJSON,
-	})
-	if err != nil {
-		return database.Collections{}, fmt.Errorf("failed to save rules: %w", err)
-	}
-	return updated, nil
-}
-
-// RemoveRule deletes a single rule from the ruleset. Removing the LAST
-// rule converts the collection back to static (type is derived).
-func (s *CollectionService) RemoveRule(ctx context.Context, collectionID uuid.UUID, ruleID string) (database.Collections, error) {
-	collection, err := s.GetCollection(ctx, collectionID)
-	if err != nil {
-		return database.Collections{}, err
-	}
-	rules := s.ParseRules(collection.AutoAssignRules)
-	kept := make([]Rule, 0, len(rules))
-	found := false
-	for _, rule := range rules {
-		if rule.ID == ruleID {
-			found = true
-			continue
-		}
-		kept = append(kept, rule)
-	}
-	if !found {
-		return database.Collections{}, fmt.Errorf("rule not found: %s", ruleID)
-	}
-
-	rulesJSON, marshalErr := json.Marshal(kept)
-	if marshalErr != nil {
-		return database.Collections{}, fmt.Errorf("failed to marshal rules: %w", marshalErr)
-	}
-
-	updated, err := s.db.UpdateCollectionRulesJSON(ctx, database.UpdateCollectionRulesJSONParams{
-		ID:              pgtype.UUID{Bytes: collectionID, Valid: true},
-		AutoAssignRules: rulesJSON,
-	})
-	if err != nil {
-		return database.Collections{}, fmt.Errorf("failed to save rules: %w", err)
-	}
-	return updated, nil
-}
-
-// SmartEvalOptions scopes a smart collection's live evaluation.
 type SmartEvalOptions struct {
 	// LibraryID, when valid, restricts matching to one library (the
 	// detail page's library switcher).
@@ -423,54 +257,98 @@ type SmartEvalOptions struct {
 	// VisibleIDs is the requesting user's visible-library set (family
 	// lens). Empty means no visible-library filter.
 	VisibleIDs []pgtype.UUID
+	// ViewerUserID scopes the candidate query's progress join — the
+	// played/unplayed rule is per-user. Zero UUID = no viewer (played
+	// rules then see "unplayed" for everything, honestly).
+	ViewerUserID pgtype.UUID
 }
 
-// EvaluateSmartCollection resolves a smart collection's membership LIVE:
+// IsCollectionSmart is the ONE smart detection: a collection is smart
-// every visible live book matched by at least one enabled rule is in.
+// when it carries the unified filter tree (item 3.2). Every surface
-// No stored links are consulted — "truly dynamic," always current.
+// routes through this. (The legacy rules-array branch retired
-func (s *CollectionService) EvaluateSmartCollection(ctx context.Context, rules []Rule, opts SmartEvalOptions) ([]database.MediaItems, error) {
+// 2026-10-31 — the tree is the only smart storage.)
-	enabled := s.EnabledRules(rules)
+func IsCollectionSmart(coll database.Collections) bool {
-	if len(enabled) == 0 {
+	return len(coll.FilterTree) > 0
-		return []database.MediaItems{}, nil
+}
-	}
 
-	var visibleArg []pgtype.UUID
+// Search exposes the composed search engine the service was built with
-	if len(opts.VisibleIDs) > 0 {
+// (nil when the constructor omitted it) — callers composing secondary
-		visibleArg = opts.VisibleIDs
+// services that evaluate trees (the dashboard service) pass it through.
-	}
+func (s *CollectionService) Search() *SearchService { return s.search }
-	items, err := s.db.GetVisibleLiveMediaForRules(ctx, database.GetVisibleLiveMediaForRulesParams{
+
-		LibraryID:         opts.LibraryID,
+// IsCollectionSmartColl is IsCollectionSmart on the service handle —
-		VisibleLibraryIds: visibleArg,
+// the dashboard service composes this service and calls the detector
-	})
+// through it (no import cycle, one detection definition).
+func (s *CollectionService) IsCollectionSmartColl(coll database.Collections) bool {
+	return IsCollectionSmart(coll)
+}
+
+// EvaluateSmartMembership is THE single smart evaluator (item 3.2):
+// tree-bearing collections evaluate through the unified filter engine.
+func (s *CollectionService) EvaluateSmartMembership(ctx context.Context, coll database.Collections, opts SmartEvalOptions) ([]database.MediaItems, error) {
+	return s.evaluateTreeMembership(ctx, coll, opts)
+}
+
+func (s *CollectionService) evaluateTreeMembership(ctx context.Context, coll database.Collections, opts SmartEvalOptions) ([]database.MediaItems, error) {
+	tree, err := ParseFilterTree(string(coll.FilterTree))
 	if err != nil {
-		return nil, fmt.Errorf("failed to load candidate books: %w", err)
+		return nil, fmt.Errorf("collection %s has an invalid filter tree: %v", coll.ID, err)
 	}
-
+	if s.search == nil {
-	matches := make([]database.MediaItems, 0, len(items))
+		return nil, errors.New("tree collections require the search service (constructor wiring)")
-	for _, item := range items {
-		if s.mediaItemMatchesAnyRule(item, enabled) {
-			matches = append(matches, item)
 	}
-	}
+	return s.search.ComposedLiveMedia(ctx, opts.ViewerUserID, opts.LibraryID, opts.VisibleIDs, tree)
-	return matches, nil
 }
 
-func (s *CollectionService) mediaItemMatchesAnyRule(item database.MediaItems, rules []Rule) bool {
+// episodeFinished is THE canonical finished definition (same predicate
-	for _, rule := range rules {
+// as hide_finished): viewer progress percentage >= 0.999.
-		if s.mediaItemMatchesRule(item, rule) {
+func episodeFinished(viewerPercentage float64) bool {
+	return viewerPercentage >= 0.999
+}
+
+// FieldMatches is THE single rule evaluator (the rules-builder preview
+// delegates here too — one vocabulary, one semantics). Episode/user-state
+// fields (feed/played/downloaded/published_within) evaluate directly;
+// everything else extracts a string and hits the generic comparator.
+func (s *CollectionService) FieldMatches(item database.MediaItems, viewerPercentage float64, field, operator, value string) bool {
+	switch field {
+	case "feed":
+		if !item.SourceFeedID.Valid {
+			return false
+		}
+		feedID := uuid.UUID(item.SourceFeedID.Bytes).String()
+		for _, raw := range strings.Split(value, ",") {
+			if strings.TrimSpace(raw) == feedID {
 				return true
 			}
 		}
 		return false
-}
+	case "played":
-
+		wantPlayed := value == "true"
-func (s *CollectionService) mediaItemMatchesRule(item database.MediaItems, rule Rule) bool {
+		return episodeFinished(viewerPercentage) == wantPlayed
-	if rule.Enabled != nil && !*rule.Enabled {
+	case "downloaded":
+		wantDownloaded := value == "true"
+		return (item.DownloadState == "downloaded") == wantDownloaded
+	case "published_within":
+		if !item.DatePublished.Valid {
 			return false
 		}
+		days, err := strconv.Atoi(strings.TrimSpace(value))
+		if err != nil || days < 0 {
+			return false
+		}
+		cutoff := time.Now().UTC().AddDate(0, 0, -days)
+		return item.DatePublished.Time.After(cutoff)
+	}
 
 	var fieldValue string
-	switch rule.Field {
+	switch field {
+	case "title":
+		fieldValue = item.Title
+	case "description":
+		if item.Description.Valid {
+			fieldValue = item.Description.String
+		}
 	case "genre":
 		if item.Genre.Valid {
 			fieldValue = item.Genre.String
@@ -501,11 +379,14 @@ func (s *CollectionService) mediaItemMatchesRule(item database.MediaItems, rule
 		}
 	}
 
-	return s.evaluateRule(fieldValue, rule.Operator, rule.Value)
+	return s.evaluateRule(fieldValue, operator, value)
 }
 
-// EvaluateRules evaluates auto-assign rules for a media item
+// EvaluateRules evaluates auto-assign rules for a media item — the
-func (s *CollectionService) EvaluateRules(mediaItem database.MediaItems, rules []Rule) []RuleEvaluation {
+// rules-builder preview's per-rule view. viewerPercentage is the
+// requesting user's progress for this item (0 = none), so played /
+// unplayed rules preview honestly (plan item 2.4).
+func (s *CollectionService) EvaluateRules(mediaItem database.MediaItems, viewerPercentage float64, rules []Rule) []RuleEvaluation {
 	var evaluations []RuleEvaluation
 
 	// Sort rules by priority (higher priority first)
@@ -527,40 +408,7 @@ func (s *CollectionService) EvaluateRules(mediaItem database.MediaItems, rules [
 		}
 		eval := RuleEvaluation{
 			RuleID:  rule.ID,
-			Matches: false,
+			Matches: s.FieldMatches(mediaItem, viewerPercentage, rule.Field, rule.Operator, rule.Value),
-		}
-
-		switch rule.Field {
-		case "genre":
-			if mediaItem.Genre.Valid {
-				eval.Matches = s.evaluateRule(mediaItem.Genre.String, rule.Operator, rule.Value)
-			}
-		case "series":
-			if mediaItem.Series.Valid {
-				eval.Matches = s.evaluateRule(mediaItem.Series.String, rule.Operator, rule.Value)
-			}
-		case "author":
-			if mediaItem.Author.Valid {
-				eval.Matches = s.evaluateRule(mediaItem.Author.String, rule.Operator, rule.Value)
-			}
-		case "language":
-			if mediaItem.Language.Valid {
-				eval.Matches = s.evaluateRule(mediaItem.Language.String, rule.Operator, rule.Value)
-			}
-		case "publisher":
-			if mediaItem.Publisher.Valid {
-				eval.Matches = s.evaluateRule(mediaItem.Publisher.String, rule.Operator, rule.Value)
-			}
-		case "copyright_year":
-			if mediaItem.CopyrightYear.Valid {
-				yearStr := fmt.Sprintf("%d", mediaItem.CopyrightYear.Int32)
-				eval.Matches = s.evaluateRule(yearStr, rule.Operator, rule.Value)
-			}
-		case "tags":
-			if len(mediaItem.Tags) > 0 {
-				tagsStr := strings.Join(mediaItem.Tags, ", ")
-				eval.Matches = s.evaluateRule(tagsStr, rule.Operator, rule.Value)
-			}
 		}
 
 		// Calculate confidence based on rule type and priority
@@ -654,6 +502,121 @@ func compareNumbers(a, b, op string) bool {
 }
 
 // CreateDeviceShelfMapping creates a shelf mapping for a device
+
+// ── Custom-section-builder adapter (2026-10-31) ──────────────────
+// The builder posts the legacy rules vocabulary; these convert it to a
+// qf tree at WRITE time — one-way upgrade, storage stays tree-only.
+// (The 2026-10-31 retirement deleted these as READ-time conversion;
+// the builder's tree migration revived them as the single write path.)
+
+// The builder's match_type select: "any" (OR) | "all" (AND).
+const (
+	RuleMatchAny = "any"
+	RuleMatchAll = "all"
+)
+
+// The builder posts the legacy rules vocabulary; these convert it to a
+// qf tree at WRITE time — one-way upgrade, storage stays tree-only.
+// (The 2026-10-31 retirement deleted these as READ-time conversion;
+// the builder's tree migration revived them as the single write path.)
+
+// RulesToFilterTree converts a legacy rules array + its match mode
+// into a qf tree so a collection upgrades to the unified engine at its
+// first edit (item 3.2): zero-rewrite back-compat at read, one-way
+// upgrade at write. Episode rules map mechanically (feed→in-list,
+// played/downloaded→booleans, published_within→published_date
+// within_days); legacy operators without an engine equivalent
+// (starts_with/ends_with) fall back to contains — documented, and the
+// operator labels stay in the rules docs.
+func RulesToFilterTree(rules []Rule, matchMode string) (*FilterTree, error) {
+	children := make([]FilterChild, 0, len(rules))
+	scope := ""
+	for _, r := range rules {
+		child, err := ruleToFilterChild(r)
+		if err != nil {
+			return nil, err
+		}
+		if episodeOnlyFields[r.Field] {
+			scope = ScopeEpisodes
+		}
+		children = append(children, child)
+	}
+	if len(children) == 0 {
+		return nil, fmt.Errorf("no rules to convert")
+	}
+	mode := RuleMatchAny
+	if matchMode == RuleMatchAll {
+		mode = RuleMatchAll
+	}
+	return &FilterTree{Scope: scope, Mode: mode, Children: children}, nil
+}
+
+func ruleToFilterChild(r Rule) (FilterChild, error) {
+	child := FilterChild{Field: r.Field}
+	switch r.Field {
+	case "feed":
+		// The rules format stores comma-separated feed IDs in one value.
+		child.Op = OpIn
+		child.Values = splitRuleValues(r.Value)
+		return child, nil
+	case "downloaded":
+		child.Op = isTrueFalse(r.Value == "true")
+		return child, nil
+	case "played":
+		child.Op = isTrueFalse(r.Value == "true")
+		return child, nil
+	case "published_within":
+		child.Field = "published_date"
+		child.Op = OpWithinDays
+		child.Values = []string{r.Value}
+		return child, nil
+	}
+	// Registry fields: legacy operators → engine operators.
+	switch r.Operator {
+	case "equals":
+		child.Op = OpIs
+	case "not_equals":
+		child.Op = OpIsNot
+	case "contains":
+		child.Op = OpContains
+	case "not_contains":
+		child.Op = OpNotContains
+	case "greater_than":
+		child.Op = OpAtLeast
+	case "less_than":
+		child.Op = OpAtMost
+	case "in":
+		child.Op = OpIn
+		child.Values = splitRuleValues(r.Value)
+	case "starts_with", "ends_with":
+		// No engine equivalent — contains is the honest nearest.
+		child.Op = OpContains
+	default:
+		return FilterChild{}, fmt.Errorf("legacy rule operator %q has no filter-engine equivalent", r.Operator)
+	}
+	child.Values = []string{r.Value}
+	return child, nil
+}
+
+func isTrueFalse(b bool) string {
+	if b {
+		return OpIsTrue
+	}
+	return OpIsFalse
+}
+
+func splitRuleValues(v string) []string {
+	parts := strings.Split(v, ",")
+	out := make([]string, 0, len(parts))
+	for _, p := range parts {
+		p = strings.TrimSpace(p)
+		if p != "" {
+			out = append(out, p)
+		}
+	}
+	return out
+}
+
 func (s *CollectionService) CreateDeviceShelfMapping(ctx context.Context, collectionID, deviceID uuid.UUID, deviceShelfName, syncDirection string) (database.DeviceShelfMappings, error) {
 	mapping, err := s.db.CreateDeviceShelfMapping(ctx, database.CreateDeviceShelfMappingParams{
 		CollectionID:    pgtype.UUID{Bytes: collectionID, Valid: true},

internal/services/collection_service_test.go

@@ -24,7 +24,7 @@ func TestEvaluateRules_Equality(t *testing.T) {
 		},
 	}
 
-	evaluations := service.EvaluateRules(mediaItem, rules)
+	evaluations := service.EvaluateRules(mediaItem, 0, rules)
 
 	if len(evaluations) != 1 {
 		t.Fatalf("expected 1 evaluation, got %d", len(evaluations))
@@ -56,7 +56,7 @@ func TestEvaluateRules_NoMatch(t *testing.T) {
 		},
 	}
 
-	evaluations := service.EvaluateRules(mediaItem, rules)
+	evaluations := service.EvaluateRules(mediaItem, 0, rules)
 
 	if len(evaluations) != 1 {
 		t.Fatalf("expected 1 evaluation, got %d", len(evaluations))
@@ -88,7 +88,7 @@ func TestEvaluateRules_Contains(t *testing.T) {
 		},
 	}
 
-	evaluations := service.EvaluateRules(mediaItem, rules)
+	evaluations := service.EvaluateRules(mediaItem, 0, rules)
 
 	if len(evaluations) != 1 {
 		t.Fatalf("expected 1 evaluation, got %d", len(evaluations))
@@ -128,7 +128,7 @@ func TestEvaluateRules_PriorityOrder(t *testing.T) {
 		},
 	}
 
-	evaluations := service.EvaluateRules(mediaItem, rules)
+	evaluations := service.EvaluateRules(mediaItem, 0, rules)
 
 	if len(evaluations) != 2 {
 		t.Fatalf("expected 2 evaluations, got %d", len(evaluations))
@@ -164,7 +164,7 @@ func TestEvaluateRules_CopyrightYear(t *testing.T) {
 		},
 	}
 
-	evaluations := service.EvaluateRules(mediaItem, rules)
+	evaluations := service.EvaluateRules(mediaItem, 0, rules)
 
 	if len(evaluations) != 1 {
 		t.Fatalf("expected 1 evaluation, got %d", len(evaluations))
@@ -190,7 +190,7 @@ func TestEvaluateRules_MissingField(t *testing.T) {
 		},
 	}
 
-	evaluations := service.EvaluateRules(mediaItem, rules)
+	evaluations := service.EvaluateRules(mediaItem, 0, rules)
 
 	if len(evaluations) != 1 {
 		t.Fatalf("expected 1 evaluation, got %d", len(evaluations))
@@ -218,7 +218,7 @@ func TestEvaluateRules_CaseInsensitive(t *testing.T) {
 		},
 	}
 
-	evaluations := service.EvaluateRules(mediaItem, rules)
+	evaluations := service.EvaluateRules(mediaItem, 0, rules)
 
 	if len(evaluations) != 1 {
 		t.Fatalf("expected 1 evaluation, got %d", len(evaluations))
@@ -253,7 +253,7 @@ func TestEvaluateRules_PriorityBoost(t *testing.T) {
 		},
 	}
 
-	evaluations := service.EvaluateRules(mediaItem, rules)
+	evaluations := service.EvaluateRules(mediaItem, 0, rules)
 
 	if len(evaluations) != 2 {
 		t.Fatalf("expected 2 evaluations, got %d", len(evaluations))
@@ -276,3 +276,8 @@ func TestEvaluateRules_PriorityBoost(t *testing.T) {
 		t.Errorf("expected higher priority rule to be first, got %s", evaluations[0].RuleID)
 	}
 }
+
+// TestRulesToFilterTree retired 2026-10-31 with the function it tested:
+// rules-array storage is gone, the filter tree is the only smart
+// membership. (Tests of deleted behavior are deleted — the asymmetry
+// rule.)