feat(stats): add live-action library support and harden stats runtime

- Add TMDB metadata, linking, filtering, and live-action library merging
- Harden stats request validation, lifecycle handling, and compiled runtime coverage
- Fix Anki media synchronization and configuration validation

.github/workflows/package-release.yml

@@ -13,6 +13,9 @@ on:
         required: true
       APPLE_TEAM_ID:
         required: true
+      # Project TMDB key baked into release artifacts; builds stay valid without it.
+      SUBMINER_TMDB_API_KEY:
+        required: false
 
 permissions:
   contents: read
@@ -69,6 +72,8 @@ jobs:
 
       - name: Build AppImage
         run: bun run build:appimage
+        env:
+          SUBMINER_TMDB_API_KEY: ${{ secrets.SUBMINER_TMDB_API_KEY }}
 
       - name: Build unversioned AppImage
         run: |
@@ -168,6 +173,7 @@ jobs:
       - name: Build signed + notarized macOS artifacts
         run: bun run build:mac
         env:
+          SUBMINER_TMDB_API_KEY: ${{ secrets.SUBMINER_TMDB_API_KEY }}
           CSC_LINK: ${{ secrets.CSC_LINK }}
           CSC_KEY_PASSWORD: ${{ secrets.CSC_KEY_PASSWORD }}
           APPLE_ID: ${{ secrets.APPLE_ID }}
@@ -248,6 +254,8 @@ jobs:
 
       - name: Build unsigned Windows artifacts
         run: bun run build:win:unsigned
+        env:
+          SUBMINER_TMDB_API_KEY: ${{ secrets.SUBMINER_TMDB_API_KEY }}
 
       - name: Smoke packaged runtime assets
         shell: bash

.github/workflows/prerelease.yml

@@ -27,6 +27,7 @@ jobs:
       APPLE_ID: ${{ secrets.APPLE_ID }}
       APPLE_APP_SPECIFIC_PASSWORD: ${{ secrets.APPLE_APP_SPECIFIC_PASSWORD }}
       APPLE_TEAM_ID: ${{ secrets.APPLE_TEAM_ID }}
+      SUBMINER_TMDB_API_KEY: ${{ secrets.SUBMINER_TMDB_API_KEY }}
 
   release:
     needs: [package]

.github/workflows/quality-gate.yml

@@ -93,12 +93,20 @@ jobs:
           sudo ln -sf /usr/bin/lua5.4 /usr/local/bin/lua
           lua -v
 
-      - name: Test suite (source)
-        run: bun run test:fast
+      - name: Launcher unit and script suites
+        run: bun run test:launcher:unit:src && bun run test:scripts
 
       - name: Environment suite
         run: bun run test:env
 
+      - name: Upload launcher smoke artifacts (on failure)
+        if: failure()
+        uses: actions/upload-artifact@v4
+        with:
+          name: launcher-smoke
+          path: .tmp/launcher-smoke/**
+          if-no-files-found: ignore
+
       - name: Coverage suite (maintained source lane)
         run: bun run test:coverage:src
 
@@ -112,17 +120,6 @@ jobs:
       - name: Stats UI tests
         run: bun run test:stats
 
-      - name: Launcher smoke suite (source)
-        run: bun run test:launcher:smoke:src
-
-      - name: Upload launcher smoke artifacts (on failure)
-        if: failure()
-        uses: actions/upload-artifact@v4
-        with:
-          name: launcher-smoke
-          path: .tmp/launcher-smoke/**
-          if-no-files-found: ignore
-
       - name: Build (bundle)
         run: bun run build
 

.github/workflows/release.yml

@@ -28,6 +28,7 @@ jobs:
       APPLE_ID: ${{ secrets.APPLE_ID }}
       APPLE_APP_SPECIFIC_PASSWORD: ${{ secrets.APPLE_APP_SPECIFIC_PASSWORD }}
       APPLE_TEAM_ID: ${{ secrets.APPLE_TEAM_ID }}
+      SUBMINER_TMDB_API_KEY: ${{ secrets.SUBMINER_TMDB_API_KEY }}
 
   release:
     needs: [package]

changes/anime-browser.md

@@ -19,7 +19,7 @@ area: anime
 - The strip proxy retries a failed segment fetch once after a short pause and logs upstream error statuses; a host that errors on the very first fetches right after an episode resolves no longer kills the whole playback, disconnecting clients release their active upstream fetch instead of consuming the socket and bandwidth in the background, stalled response bodies retain the upstream inactivity timeout, and only origin-form requests can reach the configured local bridge.
 - The strip proxy no longer forwards `Range` headers to the bridge: ffmpeg opens every HLS segment with `Range: bytes=0-`, the bridge answers some of those with 206, and a partial response bypassed the disguise strip, so whether an episode played depended on the bridge's cache state.
 - A bridge that dies out from under the app (killed, crashed, or stopped mid-operation) no longer leaves the browser failing every request until an app restart: the exit is detected, surfaced in the status bar, and the bridge restarts on the next request.
-- "Playing" is only reported once mpv actually configures a video output; when a stream fails to decode, the browser shows mpv's error instead of claiming playback started while no window ever appeared.
+- "Playing" is only reported once mpv actually configures a video output; slow streams remain loading while mpv opens them, and failed streams show mpv's error.
 - The bridge bundle is downloaded from the newest upstream release that ships a bundle for the platform (skipping releases older than the oldest server SubMiner is known to work with), the same way Mangatan fetches it, so all four published bundles (macOS arm64 and x64, Linux x64, Windows x64) work without a maintainer hashing each one first.
 - The Extensions tab's available list has a language chip row: pick one or more languages to narrow a repository index that otherwise lists every language it knows, or "All" to clear the filter. Rows name the language ("Japanese" instead of `ja`) and the Available heading counts what the filter leaves.
 - A stream's subtitle tracks are downloaded to a temp directory and loaded into mpv as files rather than streamed from the source URL, so they can serve as the alass reference in Subsync the way Jellyfin subtitles do. The format is detected from the file's own content, a track that fails to download falls back to its URL so the episode still plays, and the directory is removed when the next episode starts or the app exits.

changes/anime-playback-loading-docs.md

@@ -0,0 +1,4 @@
+type: docs
+area: anime
+
+- Document how the anime browser waits for slow streams to start without reporting a premature playback failure.

changes/ci-test-deduplication.md

@@ -0,0 +1,4 @@
+type: internal
+area: ci
+
+- Removed duplicate source and launcher smoke executions from the reusable quality gate while preserving every distinct test lane and failure artifact.

changes/compiled-runtime-smoke.md

@@ -0,0 +1,4 @@
+type: internal
+area: verification
+
+- Replaced mislabeled dist source reruns with a small Electron-runtime smoke check for compiled stats startup, HTTP service, native SQLite, port conflicts, and cleanup.

changes/fix-animated-word-audio-sync.md

@@ -0,0 +1,4 @@
+type: fixed
+area: anki
+
+- Added `ankiConnect.fields.wordAudio` to read word audio separately from the generated sentence-audio destination, fixing animated images that start moving immediately when `fields.audio` points to `SentenceAudio`.

changes/fix-anki-config-validation.md

@@ -0,0 +1,4 @@
+type: fixed
+area: config
+
+- Validate direct AnkiConnect, Kiku, and Senren settings before admitting them to runtime config, with warnings and defaults for invalid values.

changes/fix-jellyfin-media-identity.md

@@ -0,0 +1,7 @@
+type: fixed
+area: jellyfin
+
+- Set the mpv title before loading Jellyfin streams and reject URL-derived titles from metadata lookups, Anki source fields, Discord presence, and stats.
+- Keep authenticated stream URLs out of stats identities even when playback metadata has not arrived.
+- Remove previously cached credential-bearing Jellyfin stream parser metadata without changing unrelated anime metadata, watch history, or library assignments.
+- Use safe media identities for persisted AniList retry keys, discard URL-derived queued searches, and skip updates and retries whose media identity is empty.

changes/fix-stats-resource-id-validation.md

@@ -0,0 +1,4 @@
+type: fixed
+area: stats
+
+- Reject malformed resource IDs and partly invalid ID lists before stats library mutations or cover backfills run.

changes/fix-stats-server-lifecycle.md

@@ -0,0 +1,5 @@
+type: fixed
+area: stats
+
+- Stats server startup reports port conflicts without crashing SubMiner, shares concurrent startup requests, and shows in-app startup errors through configured status notifications.
+- Background stop cancels pending background startup without disconnecting foreground-only dashboards. Shutdown bounds the wait for active HTTP requests and awaits tracker finalization before exit, with a deadline for forced application exit.

changes/fix-timing-review-cancellation.md

@@ -0,0 +1,4 @@
+type: fixed
+area: anki
+
+- Closing the overlay while media timing review is still loading now cancels setup and modal retries, restores playback if the review paused it, and cleans up the hidden preview player.

changes/fix-unlimited-mining-duration.md

@@ -0,0 +1,4 @@
+type: fixed
+area: anki
+
+- Treat `ankiConnect.media.maxMediaDuration: 0` as unlimited for stats dashboard mining, matching overlay mining and configuration.

changes/japanese-subtitle-generation-docs.md

@@ -2,3 +2,4 @@ type: docs
 area: subtitles
 
 - Explain how dialogue generation retains uncertain audible sections, why songs may also be transcribed, how detected speech starts guide long-passage cuts to reduce early subtitles, and why each passage uses a fresh Whisper process.
+- Document automatic loaded subtitle references, English dialogue preference, signs/songs filtering, fallback behavior, and the limits of reference-guided timing.

changes/japanese-subtitle-generation.md

@@ -4,3 +4,5 @@ area: subtitles
 - Generate local Japanese SRT subtitles with whisper.cpp from a standalone modal opened with Ctrl+Shift+G, the empty subtitle sidebar's generation button, or `subminer generate-subs`, with shared progress reporting, cancellation, safe output files, and automatic loading into the matching mpv video. The sidebar button hides while subtitle lines are loaded.
 - Configure an existing multilingual model in Settings or choose an official multilingual model, including quantized variants, in the modal or launcher. The modal shows download sizes, speed and accuracy guidance, and a recommended starting model before explicitly downloading a verified SubMiner-managed model. Executable paths are optional overrides; empty fields find whisper-cli, ffmpeg, and ffprobe on PATH. The modal's Local tools check and the launcher name any missing executable and its setting before downloading a model or extracting audio, and generation confirms the destination directory grants write and search permissions up front.
 - Optionally select Focus on spoken dialogue in the modal and use Download speech detection model to install the separate Silero model with progress and cancellation. The choice lasts for the session; a configured VAD model path sets the default. Retain uncertain audible sections so VAD rejection does not discard dialogue under music, accepting that songs may also be transcribed. Keep passages intact within Whisper's audio window, split longer passages near detected speech starts or quiet pauses with overlapping context to reduce early subtitle timing, and combine overlapping duplicate cues even when punctuation differs. Run each passage in a fresh Whisper process to prevent repeated-character output caused by state carried between files, at the cost of reloading the model per passage. Preserve original media timing and separate repeated dialogue.
+- Recommend large-v3-turbo in the model picker when NVIDIA hardware and CUDA device support in the selected Whisper executable are confirmed; recommend small otherwise. Preserve configured and session model selections. Use a wider generation modal to reduce text wrapping and scrolling, while keeping it within smaller player windows.
+- Automatically use an eligible embedded or external subtitle track loaded in mpv to guide long-passage cuts, preferring English dialogue and excluding tracks marked signs, songs, forced, or generated. Honor active subtitle delays, retain unreferenced audio, and fall back to audio timing when references cannot be read. Available in the overlay and launcher, with or without VAD. The launcher captures audio selection, reference tracks, and delays before model setup so playback changes during a download do not mix inputs from different times, and rejects an unverified media snapshot before using mpv's selected audio stream.

changes/jellyfin-media-identity-docs.md

@@ -0,0 +1,5 @@
+type: docs
+area: jellyfin
+
+- Explain how Jellyfin media titles, unknown-title fallbacks, and stats identities keep stream credentials out of metadata.
+- Document AniList rejection of empty media identities and the conditions for clearing cached Jellyfin parser metadata.

changes/jimaku-live-action-search.md

@@ -0,0 +1,4 @@
+type: added
+area: jimaku
+
+- Jimaku modal gains Anime / Live action tabs so searches can pull from Jimaku's live action catalogue instead of only anime entries (Arrow Left / Right switch tabs).

changes/media-review-frame-picker-docs.md

@@ -0,0 +1,4 @@
+type: docs
+area: mining
+
+- Document independent screenshot selection in media timing review.

changes/media-review-frame-picker.md

@@ -0,0 +1,4 @@
+type: added
+area: mining
+
+- Choose a still screenshot independently of the audio in media timing review, with a live preview, a separate time slider, and decoded-frame stepping. Local video and seekable remote streams such as Jellyfin are supported.

changes/runtime-ownership.md

@@ -0,0 +1,4 @@
+type: fixed
+area: overlay
+
+- Cancel pending Linux overlay window replacements during teardown so a delayed close callback cannot reopen the overlay.

changes/settings-live-save-feedback-docs.md

@@ -0,0 +1,4 @@
+type: docs
+area: config
+
+- Clarified live-setting save feedback, mixed restart warnings, and subtitle-generation reload behavior.

changes/settings-live-save-feedback.md

@@ -0,0 +1,5 @@
+type: fixed
+area: config
+
+- Settings marked LIVE now use the same reload policy as save results, fixing false restart warnings for notifications and subtitle generation.
+- Mixed saves apply live changes and list only sections with changed fields that require a restart.

changes/stats-request-safety.md

@@ -0,0 +1,6 @@
+type: changed
+breaking: true
+area: stats
+
+- Reject requests from untrusted browser origins and hosts before stats data, media, or Anki operations run, and require JSON for mutation bodies.
+- Load the in-app stats overlay from the local server so it uses the same origin protection as the browser dashboard.

changes/tmdb-linking-docs.md

@@ -0,0 +1,4 @@
+type: docs
+area: stats
+
+- Documented provider reassignment, merge compatibility, and TMDB credential command caching and retry cooldown.

changes/tmdb-live-action-library.md

@@ -0,0 +1,8 @@
+type: added
+area: stats
+
+- Live-action dramas and movies in the stats Library now get posters, synopses, and titles from TMDB. Release builds include a project key, so it works out of the box; `tmdb.apiKey` (or `tmdb.apiKeyCommand`) overrides it, and is required when running from source.
+- Unlinked titles that AniList cannot match are looked up on TMDB automatically when the parsed filename matches a Japanese live-action title exactly; otherwise use the new **Link to TMDB** action on a title to pick it by hand.
+- Entries linked to the same TMDB title are merged into one card even when they came from different season folders, and the Library kind selector gained a Live Action option alongside Anime and YouTube.
+- Provider reassignment preserves the previous link and artwork if the replacement download fails, and refreshes completion totals when the episode count changes. Merges and sync keep conflicting AniList and TMDB identities separate, and the merge dialog explains when a selection mixes the two instead of failing.
+- TMDB credential commands cache successful output and wait 30 seconds before retrying failed or empty output, using the bundled key in the meantime when available.

changes/word-audio-mapping-docs.md

@@ -0,0 +1,4 @@
+type: docs
+area: anki
+
+- Documented the separate word-audio mapping for animated-image synchronization and that existing images need regeneration to pick up the corrected freeze.

changes/yomitan-upstream-26-9-8.md

@@ -0,0 +1,4 @@
+type: changed
+area: yomitan
+
+- Updated bundled Yomitan with upstream 26.9.8 changes, including historical Japanese kana transformations, Ukrainian language support, and improvements to Anki duplicate searches and audio retrieval.

changes/youtube-library-kind-docs.md

@@ -0,0 +1,4 @@
+type: docs
+area: stats
+
+- Document YouTube channel filtering and video statistics in the Library.

changes/youtube-library-kind.md

@@ -0,0 +1,7 @@
+type: added
+area: stats
+
+- Store YouTube channels as a separate media kind and migrate existing channel entries without changing viewing history or manual video assignments, including channels created after temporarily returning to an older build.
+- Add All Titles, Anime, and YouTube Library filters, identify channel pages, and keep channels out of AniList matching, season repair, and duplicate recommendations.
+- Keep same-title anime and YouTube records separate in storage and sync while repairing legacy channel classification.
+- Refuse merges and video moves that would mix an anime entry with a YouTube channel; the move picker and merge selection only combine entries of the same kind.

config.example.jsonc

@@ -547,7 +547,7 @@
   // ==========================================
   // AnkiConnect Integration
   // Automatic Anki updates and media generation options.
-  // Hot-reload: ankiConnect.ai.enabled, media.normalizeAudio/mirrorMpvVolume/reviewTiming, knownWords, nPlusOne, fields.word/audio/image/sentence/miscInfo, behavior.autoUpdateNewCards, isLapis.sentenceCardModel, isKiku.fieldGrouping, isSenren.fieldGrouping, and lapisKiku.wordCardKind update live while SubMiner is running.
+  // Hot-reload: ankiConnect.ai.enabled, media.normalizeAudio/mirrorMpvVolume/reviewTiming, knownWords, nPlusOne, fields.word/audio/wordAudio/image/sentence/miscInfo, behavior.autoUpdateNewCards, isLapis.sentenceCardModel, isKiku.fieldGrouping, isSenren.fieldGrouping, and lapisKiku.wordCardKind update live while SubMiner is running.
   // Shared AI provider transport settings are read from top-level ai and typically require restart.
   // Most other AnkiConnect settings still require restart.
   // ==========================================
@@ -568,6 +568,7 @@
     "fields": {
       "word": "Expression", // Card field for the mined word or expression text.
       "audio": "ExpressionAudio", // Card field that receives generated sentence audio.
+      "wordAudio": "ExpressionAudio", // Existing word-audio field read to time the frozen first frame of animated images. This mapping is only used for synchronization.
       "image": "Picture", // Card field that receives the captured screenshot or animated image.
       "sentence": "Sentence", // Card field that receives the source sentence text.
       "miscInfo": "MiscInfo", // Card field that receives the miscellaneous info pattern (see ankiConnect.metadata.pattern).
@@ -596,7 +597,7 @@
       "reviewTiming": false, // Review and preview subtitle media timing before SubMiner creates or enriches a mined card. Values: true | false
       "audioPadding": 0, // Seconds of padding appended to both ends of generated sentence audio and animated AVIF clips.
       "fallbackDuration": 3, // Fallback clip duration in seconds when subtitle timing data is unavailable.
-      "maxMediaDuration": 30 // Maximum allowed media clip duration in seconds.
+      "maxMediaDuration": 30 // Maximum allowed media clip duration in seconds. 0 disables the cap.
     }, // Media setting.
     "knownWords": {
       "highlightEnabled": false, // Enable fast local highlighting for words already known in Anki. Values: true | false
@@ -679,6 +680,16 @@
     "maxSearchResults": 10 // Maximum TsukiHime search results returned.
   }, // TsukiHime subtitle search configuration for Japanese primary and configured secondary subtitles. No API key required.
 
+  // ==========================================
+  // TMDB
+  // TMDB (The Movie Database) metadata for live-action dramas and movies in the stats Library: posters, synopses, and grouping by show.
+  // Hot-reload: TMDB changes apply to the next TMDB request.
+  // ==========================================
+  "tmdb": {
+    "apiKey": "", // Your own TMDB API key or read access token for live-action posters and synopses in the stats Library. Release builds bundle a project key, so set this only to use your own quota or when running from source (free under Settings > API on themoviedb.org).
+    "apiKeyCommand": "" // Shell command that prints the TMDB API key to stdout. Used instead of apiKey to avoid storing the key in plain text.
+  }, // TMDB (The Movie Database) metadata for live-action dramas and movies in the stats Library: posters, synopses, and grouping by show.
+
   // ==========================================
   // YouTube Playback Settings
   // Defaults for managed subtitle language preferences and YouTube subtitle loading.

docs-site/anilist-integration.md

@@ -47,6 +47,8 @@ The update flow:
 
 Failed AniList updates are persisted to a retry queue on disk and retried with exponential backoff.
 
+Updates are skipped if the media path cannot produce a safe, nonempty identity. Invalid entries are discarded when loading or adding to the retry queue.
+
 | Parameter        | Value      |
 | ---------------- | ---------- |
 | Initial backoff  | 30 seconds |
@@ -69,6 +71,8 @@ SubMiner fetches cover art from AniList for display in the stats dashboard. When
 
 A no-match result is cached for 5 minutes before SubMiner retries, preventing repeated API calls for unrecognized media.
 
+When AniList has no match, SubMiner tries [TMDB](/configuration#tmdb) next so live-action dramas and movies get a poster and synopsis too. See [Immersion tracking](/immersion-tracking#library) for how live-action entries are grouped.
+
 If the automatic match is wrong, use **Change AniList Entry** on a title in the stats Library. Relinking rewrites the cached art for every episode of that title, and both the detail view and the Library grid pick up the new cover right away: the grid refetches after a relink, and cover responses carry an ETag and are revalidated on each request instead of being cached for a day.
 
 ## Rate limiting

docs-site/anime-browser.md

@@ -339,6 +339,10 @@ SubMiner waits until mpv actually configures a video output before reporting
 success. If mpv gives up instead — a dead host, an undecodable stream — the
 browser shows mpv's error rather than pretending playback started (a failed
 load leaves no mpv window, because the player idles windowless).
+Slow streams stay in the loading state while mpv is still opening them, even
+after the initial playback confirmation deadline. Loading ends when video
+appears, mpv reports an error or returns to idle, or another episode replaces
+the request.
 
 Failed requests appear in a dismissible panel at the bottom of the browser.
 It explains the failed action and suggests a next step. Expand **Technical

docs-site/anki-integration.md

@@ -19,6 +19,8 @@ This project is built primarily for [Kiku](https://kiku.youyoumu.my.id/) and [La
 
 AnkiConnect listens on `http://127.0.0.1:8765` by default. If you changed the port in AnkiConnect's settings, update `ankiConnect.url` in your SubMiner config.
 
+AnkiConnect and Kiku/Senren settings follow the [configuration validation rules](/configuration#configuration-file): invalid values produce a warning and fall back to the option's default. Use JSON booleans such as `true`, not strings such as `"true"`, and a positive number for `ankiConnect.pollingRate`.
+
 ## Auto-enrichment transport
 
 When you add a word via Yomitan, SubMiner detects the new card and fills in the sentence, audio, and image fields automatically. Two detection methods are available:
@@ -125,6 +127,7 @@ SubMiner maps its data to your Anki note fields. Configure these under `ankiConn
   "fields": {
     "word": "Expression",        // mined word / expression text
     "audio": "SentenceAudio",    // sentence audio clip cut from the video
+    "wordAudio": "ExpressionAudio", // existing Yomitan word audio, read for animation sync
     "image": "Picture",          // screenshot or animated clip
     "sentence": "Sentence",      // subtitle text
     "miscInfo": "MiscInfo"       // metadata (filename, timestamp)
@@ -136,6 +139,8 @@ SubMiner maps its data to your Anki note fields. Configure these under `ankiConn
 
 Field names are matched against your Anki note type case-insensitively (an exact match wins, then a lowercase comparison). If a configured field does not exist on the note type, SubMiner skips it without error.
 
+`fields.wordAudio` selects the existing dictionary-audio field used to calculate the animated image's opening freeze. This mapping only reads audio; `fields.audio` still controls where generated sentence audio is written. See [config.example.jsonc](/config.example.jsonc) for defaults.
+
 These mappings always control normal word-card enrichment, including Yomitan proxy/polling updates and manual clipboard updates. Enabling Lapis or Kiku does not replace the configured word-card sentence and audio fields with `Sentence` and `SentenceAudio`. The dedicated sentence-card and audio-card shortcuts still use those Lapis/Kiku field names.
 
 Two related options live alongside `fields`: `ankiConnect.deck` (target deck; empty falls back as described above) and `ankiConnect.tags` (tags added to mined cards, default `["SubMiner"]`; set `[]` to disable tagging). The `miscInfo` content is controlled by `ankiConnect.metadata.pattern` (default `[SubMiner] %f (%t)`; tokens: `%f` filename, `%F` filename with extension, `%t` timestamp, `%T` timestamp with milliseconds, `<br>` newline).
@@ -183,12 +188,18 @@ Output format: MP3 at 44100 Hz. If the video has multiple audio streams, SubMine
 
 The audio is uploaded to Anki's media folder and inserted as `[sound:audio_<timestamp>.mp3]`.
 
+Overlay and stats-dashboard mining use the same `media.maxMediaDuration` limit. See the [configuration example](/config.example.jsonc) for its default and how to disable the cap.
+
 Set `media.reviewTiming` to `true` to pause playback and check the clip before its media is generated. It applies to word, sentence, and audio cards.
 
 The review opens on the subtitle range plus your configured audio padding. Subtitles usually hang around after the dialogue has stopped, so once the waveform loads, an untouched clip end pulls back to just after the last speech in the line. The Line end rail still marks the original subtitle timing, Reset puts it back, and a line whose speech runs right through its end is left alone.
 
 **Adjusting the clip.** Drag either edge to trim, drag the middle to slide the whole clip without changing its length, or click anywhere on the waveform to snap the nearer edge there. A focused edge also moves with the arrow keys: 100 ms per press, or 500 ms with Shift. The 100 ms buttons do the same thing. Earlier and Later each reveal two more seconds of timeline without moving the selection.
 
+**Choosing the screenshot.** With still images enabled, drag the screenshot slider or use the Frame buttons to choose a video frame independently of the audio. Earlier and Later reveal more time for both sliders. The image follows the audio midpoint until you pick a frame, then stays fixed while you trim audio. Reset below the image restores automatic selection; the audio Reset affects only the audio. The screenshot slider also supports arrow keys, Home, and End.
+
+The picker supports local video and seekable remote streams, including Jellyfin, without seeking the main player. A selected frame must load before you can confirm; if it fails, choose another frame or Reset. The picker is hidden when image generation is disabled or animated AVIF is selected. Animated images continue to use the reviewed audio range.
+
 **Keys.** Space previews the selection with a playhead sweeping the clip. The preview ends when the hidden player has actually played the last sample, so Bluetooth output latency does not clip the tail. Enter confirms and Escape cancels.
 
 **The waveform.** SubMiner reads a center channel when one carries dialogue and falls back to a mono mix otherwise, keeps only the 250 to 3500 Hz speech band, and draws each slice's loudness against the clip's own noise floor. Steady background music flattens out and dialogue stands up, which makes it much easier to tell adjacent lines apart. The mined subtitle appears as a tinted band with labeled line-start and line-end rails. If waveform analysis fails, the timing controls still work.
@@ -203,6 +214,8 @@ Confirming writes the combined lines to the sentence field. Reset drops the adde
 
 Clipboard updates and stats-dashboard mining never open timing review. The option is off by default and hot-reloads. **Review Media Timing** in the runtime options palette (`Ctrl/Cmd+Shift+O`) toggles it for the current session.
 
+If SubMiner closes the overlay while a timing review is still loading, it cancels pending setup and modal retries and restores playback if the review paused it. A new timing review can start after the overlay reopens.
+
 ### Screenshots (static)
 
 A single frame is captured at the current playback position.
@@ -237,7 +250,7 @@ SubMiner can produce an animated AVIF spanning the subtitle duration instead of
 }
 ```
 
-Animated AVIF requires an AV1 encoder (`libaom-av1`, `libsvtav1`, or `librav1e`) in your FFmpeg build. Generation timeout is 60 seconds. `media.syncAnimatedImageToWordAudio` (default `true`) prepends a frozen first frame matching the existing word-audio duration, so the motion starts together with the sentence audio.
+Animated AVIF requires an AV1 encoder (`libaom-av1`, `libsvtav1`, or `librav1e`) in your FFmpeg build. Generation timeout is 60 seconds. `media.syncAnimatedImageToWordAudio` (default `true`) prepends a frozen first frame matching the existing audio duration in `fields.wordAudio`, so the motion starts together with the sentence audio. The freeze is baked into the image when mined; changing the mapping does not repair previously generated images.
 
 ### Behavior options
 

docs-site/configuration.md

@@ -61,7 +61,7 @@ The Settings window preserves existing JSONC comments, trailing commas, and unre
 
 Secret fields do not display stored values. They show whether a value is configured; entering a new value writes it, and reset clears the explicit path. Prefer command-based secret options such as `jimaku.apiKeyCommand` when available.
 
-Saving validates the candidate config before writing. Live-reloadable changes are applied immediately; other changes return a restart-required banner in the window.
+Saving validates the candidate config before writing. Saving only fields marked **LIVE** shows "Saved. Live settings applied." If a save also changes fields that need a restart, the banner lists only the sections containing those changed fields. Live changes still apply in the same save.
 
 ## Configuration file
 
@@ -103,7 +103,7 @@ SubMiner watches the active config file (`config.jsonc` or `config.json`) while
 
 Hot-reloadable settings include subtitle appearance, sidebar controls, keybindings,
 shortcuts, notifications, logging level, selected source-language preferences,
-Jimaku/Subsync settings, AniSkip settings (`mpv.aniskipEnabled`, `mpv.aniskipButtonKey`),
+Jimaku/Subsync and subtitle-generation settings, AniSkip settings (`mpv.aniskipEnabled`, `mpv.aniskipButtonKey`),
 stats keys (`stats.toggleKey`, `stats.markWatchedKey`), the secondary-subtitle default
 mode, and the Anki deck, known-word, N+1, field, sentence-card, and Kiku options
 listed in the reference tables below.
@@ -157,6 +157,7 @@ The configuration file includes several main sections:
 - [**Anime Browser**](#anime-browser) - Extension repositories and stream preferences for the anime browser
 - [**Jimaku**](#jimaku) - Jimaku API configuration and defaults
 - [**TsukiHime**](#tsukihime) - Multi-language subtitle search and download
+- [**TMDB**](#tmdb) - Posters and synopses for live-action dramas and movies in the stats Library
 - [**Subtitle Sync**](#subtitle-sync) - Sync current subtitle with `alass`/`ffsubsync`
 - [**AniList**](#anilist) - Optional post-watch progress updates
 - [**Yomitan**](#yomitan) - Reuse an external read-only Yomitan profile
@@ -978,6 +979,7 @@ This example is intentionally compact. The option table below documents availabl
 | `ankiConnect.deck`                                | string                                      | Restrict duplicate detection and card enrichment to this Anki deck. Leave empty to use the Yomitan mining deck when available. In Settings, this dropdown auto-fills and persists Yomitan's current mining deck when available. |
 | `fields.word`                                     | string                                      | Card field for mined word / expression text (default: `Expression`)                                                                                                                                                             |
 | `fields.audio`                                    | string                                      | Card field for the generated sentence audio clip (default: `ExpressionAudio`). Set this to a dedicated field such as `SentenceAudio` so it does not collide with the word audio Yomitan writes.                                  |
+| `fields.wordAudio`                                | string                                      | Existing word-audio field read for the animated image's opening freeze. Independent of the sentence-audio destination in `fields.audio`; this mapping does not write audio. See [config.example.jsonc](/config.example.jsonc) for defaults. |
 | `fields.image`                                    | string                                      | Card field for images (default: `Picture`)                                                                                                                                                                                      |
 | `fields.sentence`                                 | string                                      | Card field for sentences (default: `Sentence`)                                                                                                                                                                                  |
 | `fields.miscInfo`                                 | string                                      | Card field for metadata (default: `"MiscInfo"`, set to `null` to disable)                                                                                                                                                       |
@@ -998,7 +1000,7 @@ This example is intentionally compact. The option table below documents availabl
 | `media.syncAnimatedImageToWordAudio`              | `true`, `false`                             | Whether animated AVIF includes an opening frame synced to sentence word-audio timing (default: `true`).                                                                                                                         |
 | `media.audioPadding`                              | number (seconds)                            | Optional padding around generated sentence media timing (default: `0`). Animated AVIF clips include the same padded source range as sentence audio.                                                                             |
 | `media.fallbackDuration`                          | number (seconds)                            | Default duration if timing unavailable (default: `3.0`)                                                                                                                                                                         |
-| `media.maxMediaDuration`                          | number (seconds)                            | Max duration for generated media from multi-line copy (default: `30`, `0` to disable)                                                                                                                                           |
+| `media.maxMediaDuration`                          | number (seconds)                            | Maximum generated clip duration for overlay and stats-dashboard mining. See the [configuration example](/config.example.jsonc) for the default and disabling the cap.                                                           |
 | `behavior.overwriteAudio`                         | `true`, `false`                             | Replace existing audio on updates; when `false`, new audio is appended/prepended using the configured media insert mode; manual clipboard updates always replace generated sentence audio (default: `true`)                     |
 | `behavior.overwriteImage`                         | `true`, `false`                             | Replace existing images on updates; when `false`, new images are appended/prepended using the configured media insert mode (default: `true`)                                                                                    |
 | `behavior.mediaInsertMode`                        | `"append"`, `"prepend"`                     | Where to insert new media when overwrite is off (default: `"append"`)                                                                                                                                                           |
@@ -1188,6 +1190,32 @@ The keyboard shortcut lives under `shortcuts.openTsukihime` (default `Ctrl+Shift
 
 See [TsukiHime Integration](/tsukihime-integration) for the modal workflow, language tabs, and troubleshooting.
 
+### TMDB
+
+TMDB (The Movie Database) supplies posters, synopses, and show grouping for live-action dramas and movies in the stats [Library](/immersion-tracking#library). AniList only covers anime, so TMDB is what gives live-action titles a cover and a description.
+
+Release builds ship with a project TMDB key, so nothing needs to be configured. Set your own key to use your own quota, or when running SubMiner from source, where no key is bundled. Create one for free under **Settings > API** on [themoviedb.org](https://www.themoviedb.org/settings/api); either the short API key or the long "API Read Access Token" works.
+
+```json
+{
+  "tmdb": {
+    "apiKey": "",
+    "apiKeyCommand": "cat ~/.tmdb_key"
+  }
+}
+```
+
+| Option               | Values | Description                                                                                        |
+| -------------------- | ------ | -------------------------------------------------------------------------------------------------- |
+| `tmdb.apiKey`        | string | Your own TMDB API key or read access token; overrides the bundled key (default: empty)             |
+| `tmdb.apiKeyCommand` | string | Shell command that prints the key to stdout, used instead of `apiKey` to keep it out of the config |
+
+Successful `apiKeyCommand` output is cached for the running client until `tmdb.apiKey` or `tmdb.apiKeyCommand` changes. Failed or empty command output uses the bundled key when available and waits 30 seconds before the next request can retry the command. Changing either credential setting resets this cooldown.
+
+Changes apply to the next TMDB request without a restart.
+
+This product uses the TMDB API but is not endorsed or certified by TMDB.
+
 ### Japanese subtitle generation
 
 Open the standalone modal with `Ctrl+Shift+G`, configurable through `shortcuts.openSubtitleGeneration`, or use the subtitle sidebar button. See [shortcuts](/shortcuts) for the shared mpv and overlay keybindings.

docs-site/immersion-tracking.md

@@ -31,12 +31,42 @@ Episode completion for local `watched` state uses the shared `DEFAULT_MIN_WATCH_
 
 The same immersion data powers the stats dashboard.
 
+The browser dashboard and in-app stats overlay both load from the local HTTP server.
+The server accepts loopback hosts only and rejects requests from other browser origins,
+including opaque origins such as `file://`. API clients without a browser origin can
+still use the local API. Mutation requests with a body must use `application/json`;
+bodyless deletion and Anki browse requests remain supported. Requests rejected by the
+host or origin checks receive `403`; mutation bodies without a JSON content type
+receive `415`.
+
+Use the loopback dashboard URL directly. Reverse-proxied dashboards and Tailscale
+Serve URLs are unsupported because their host or browser origin is not the local
+server's origin. SSH stats synchronization is unchanged.
+
+Scripts sending a JSON body must include the content type. For example, this
+requests a duplicate-line cleanup preview without changing the database. Replace
+the port if you configured a different `stats.serverPort`:
+
+```bash
+curl http://127.0.0.1:6969/api/stats/maintenance/duplicate-lines \
+  -H 'Content-Type: application/json' \
+  -d '{"dryRun":true}'
+```
+
 - In-app overlay: focus the visible overlay, then press the key from `stats.toggleKey` (default: `` ` `` / `Backquote`).
 - Launcher command: run `subminer stats` to start the local stats server on demand (it also opens the dashboard in your browser when `stats.autoOpenBrowser` is enabled; the default is `false`).
 - Background server: run `subminer stats -b` to start or reuse a dedicated background stats daemon without keeping the launcher attached, and `subminer stats -s` to stop that daemon.
 - Maintenance commands: run `subminer stats cleanup` or `subminer stats cleanup -v` to backfill/repair vocabulary metadata (`headword`, `reading`, POS) and purge stale or excluded rows from `imm_words` on demand; `subminer stats cleanup -l` repairs lifetime summary tables non-destructively (recomputed from per-episode history, so lifetime totals older than the session retention window are kept); `subminer stats cleanup --duplicate-lines` collapses repeated lines left behind by typeset subtitles (see [Repeated Line Cleanup](#repeated-line-cleanup)). `subminer stats rebuild` and `subminer stats backfill` rebuild or backfill rollup data.
 - Browser page: open `http://127.0.0.1:6969` directly if the local stats server is already running.
 
+SubMiner waits for the local server to bind before reporting that the dashboard is available. If another process already uses the configured port, the command reports the startup error and the desktop app stays open. Opening the in-app dashboard also reports startup failures through your configured status notifications.
+
+`subminer stats -s` stops a background stats server or cancels a pending background start. It leaves a foreground-only server running, so an open in-app dashboard stays connected. Shutdown gives active HTTP requests one second to finish before closing their connections and finalizing stats.
+
+### Stats API resource IDs
+
+Resource IDs in URLs must be positive safe integers written as decimal digits without leading zeros, fractions, or exponent notation. ID lists in JSON bodies must contain positive safe integer numbers. Invalid IDs or list entries return `400` before any mutation; bulk requests do not apply just the valid subset. Pagination limits keep their existing rounding and bounds.
+
 ### Dashboard tabs
 
 #### Overview
@@ -53,14 +83,16 @@ Local files and Jellyfin items with detected season numbers are split into seaso
 
 When older stats already grouped multiple seasons under one series entry, SubMiner moves parsed episodes into the season-specific entries on startup and rebuilds the affected summaries.
 
+**Live-action dramas and movies.** Anime covers come from AniList, which has no live-action titles. A title that AniList cannot match is looked up on [TMDB](/configuration#tmdb) instead (release builds bundle a key; source builds need your own): only a Japanese-language, non-animated result whose known titles match the parsed filename exactly is accepted, and it supplies the poster, synopsis, English and Japanese titles, and episode count. If nothing matches automatically, open the title and use **Link to TMDB** to search and pick it by hand. A TMDB show spans all of its seasons, so entries that resolve to the same TMDB title are merged into one card regardless of the season folder they came from, and the merged season titles are remembered so later episodes land on the same card. The **All Titles** / **Anime** / **Live Action** / **YouTube** selector above the grid narrows the Library to one kind, and a title's detail view shows whether it is a drama or a movie. Linking a title to AniList again turns it back into an anime entry. Changing providers downloads the replacement cover before saving the new link; a failed download leaves the previous link and artwork intact. A title without a cover clears the previous artwork. Automatic TMDB matching leaves existing AniList links unchanged.
+
 Jellyfin stream URLs are normalized to stable item links before stats titles are shown, so playback query parameters are not displayed in the dashboard.
 
-When YouTube channel metadata is available, the Library tab groups videos by creator/channel and treats each tracked video as an episode-like entry inside that channel section.
+When YouTube channel metadata is available, the Library tab groups videos by creator/channel. Use the kind selector above the grid (**All Titles**, **Anime**, **Live Action**, **YouTube**) to filter the library. Channel pages show tracked videos and their stats without AniList controls. Existing channel entries are classified as YouTube automatically on startup, preserving viewing history and manual video assignments. Anime and YouTube entries with the same normalized title remain separate, including during stats sync. Channels are excluded from anime metadata matching, season repair, and duplicate recommendations.
 
 A library entry is identified by its parsed title plus any detected season, so the same show can end up on several cards when releases disagree about the title or omit the season tag. Two fixes are available:
 
-- **Merge duplicates.** Hit **Select** above the grid, tick the cards that are the same show, and choose **Merge Selected**. Pick which entry to keep in the dialog; every episode moves onto it and the other cards are removed. Nothing is deleted, so sessions, mined cards and watch time all carry over. SubMiner remembers the merged title variants, so future episodes parsed with one of those names join the kept entry instead of recreating a duplicate card.
-- **Move a single episode.** Hover an episode row in a title's episode list and use the **→** button to reassign it to another library entry. The correction is remembered, so later filename parsing or Jellyfin metadata cannot move that episode back. For local files, later episodes in the same directory inherit the correction when their detected seasons are compatible and every manual correction there points to the same entry; a file that parses to a title which already has its own library entry keeps that identity instead. Conflicting seasons or manual destinations are left for review. If the move empties the old entry, that card is removed and you are returned to the grid.
+- **Merge duplicates.** Hit **Select** above the grid, tick the cards that are the same show, and choose **Merge Selected**. Pick which entry to keep in the dialog; every episode moves onto it and the other cards are removed. Nothing is deleted, so sessions, mined cards and watch time all carry over. AniList-linked and TMDB-linked entries cannot be merged together, and YouTube channels cannot be merged with anime or live-action entries. SubMiner remembers the merged title variants, so future episodes parsed with one of those names join the kept entry instead of recreating a duplicate card.
+- **Move a single episode.** Hover an episode row in a title's episode list and use the **→** button to reassign it to another library entry. Anime and live-action entries are interchangeable here, but a YouTube video can only move between channels. The correction is remembered, so later filename parsing or Jellyfin metadata cannot move that episode back. For local files, later episodes in the same directory inherit the correction when their detected seasons are compatible and every manual correction there points to the same entry; a file that parses to a title which already has its own library entry keeps that identity instead. Conflicting seasons or manual destinations are left for review. If the move empties the old entry, that card is removed and you are returned to the grid.
 
 Once cover art resolves a series to an AniList entry, cards with compatible seasons are folded together automatically only when the searched title exactly matches an AniList title or synonym. A fuzzy result that points at an AniList entry already used by another card appears as a **Possible duplicate** review above the Library grid instead. Choose **Review merge** to compare the cards and pick which one to keep, or **Not duplicates** to dismiss that suggestion permanently. Entries with conflicting explicit season numbers are left alone rather than merged or suggested.
 
@@ -327,14 +359,14 @@ LIMIT ?;
 - Large-table reads are index-backed for `sample_ms`, session time windows, frequency-ranked words/kanji, and cover-art identity lookups.
 - Workload-dependent tuning knobs remain at defaults unless you change them: `cache_size`, `mmap_size`, `temp_store`, `auto_vacuum`.
 
-### Schema (v23)
+### Schema (v24)
 
 The exact schema version lives in `SCHEMA_VERSION` (`src/core/services/immersion-tracker/types.ts`) and is recorded in the `imm_schema_version` table.
 
 Core tables:
 
 - `imm_videos` - video key/title/source metadata
-- `imm_anime` - anime/series metadata referenced by videos and lifetime tables
+- `imm_anime` - series or YouTube channel metadata referenced by videos and lifetime tables, including the media kind (`anime`, `live_action` or `youtube`) and the AniList or TMDB link
 - `imm_anime_title_aliases` - alternate titles that resolve to the same anime row
 - `imm_anime_merge_recommendations` - candidate duplicate-series merges surfaced in the dashboard
 - `imm_sessions` - session UUID, video reference, timing/status, final denormalized totals

docs-site/jellyfin-integration.md

@@ -52,10 +52,13 @@ From then on, pause / resume / seek / stop and audio or subtitle track changes y
 - **mpv launches automatically.** If mpv isn't already running when you cast, SubMiner starts it with SubMiner defaults and the bundled mpv plugin, so keybindings work right away.
 - **The overlay is managed by SubMiner,** so your configured `subtitleStyle` controls how subtitles look. Use the [overlay-toggle shortcut](/shortcuts) to hide it for a session.
 - **Resume works.** If Jellyfin has a saved position for the item, SubMiner seeks there on load.
+- **Titles and credentials stay separate.** AniList, character dictionaries, Anki source fields, and Discord presence use media titles, never authenticated stream URLs. If a usable title is unavailable, lookups are skipped and source fields show an unknown-media label. Stats identifies Jellyfin videos by server and item ID without the stream URL or API key.
 - **Direct play first.** When the source allows it and the container is in your direct-play allowlist, SubMiner streams the original file; otherwise it requests a transcoded stream from Jellyfin.
 - **Japanese subtitles are auto-selected,** preferring Jellyfin's default and embedded tracks over external sidecar files when several match.
 - **Downloaded subtitles keep their original timing.** SubMiner removes Jellyfin's server-selected subtitle stream from the mpv load URL, suppresses the mpv plugin's one-shot subtitle auto-selection and overlay auto-start for managed Jellyfin loads, stages the subtitle files exposed by Jellyfin without letting mpv auto-switch between tracks, resets mpv's subtitle delay to zero, then selects the Japanese track. SubMiner does not compare Japanese and English cue timelines or save an inferred delay.
 
+On startup, SubMiner clears cached anime parser metadata containing both API-key text and Jellyfin stream markers. Metadata containing only one of these is preserved.
+
 ## Settings
 
 All Jellyfin options live under **Settings → Integrations → Jellyfin** (open settings from the tray's **Open SubMiner Settings**). The ones that matter for casting:

docs-site/jimaku-integration.md

@@ -1,6 +1,6 @@
 # Jimaku integration
 
-[Jimaku](https://jimaku.cc) is a community subtitle repository for anime, built from files other learners uploaded. SubMiner talks to the Jimaku API, so you search, browse, and download Japanese subtitle files from inside the overlay. No alt-tabbing, no moving files around. A downloaded track loads into mpv right away.
+[Jimaku](https://jimaku.cc) is a community subtitle repository for anime and Japanese live action, built from files other learners uploaded. SubMiner talks to the Jimaku API, so you search, browse, and download Japanese subtitle files from inside the overlay. No alt-tabbing, no moving files around. A downloaded track loads into mpv right away.
 
 ::: tip Prerequisite: a free API key
 You need a Jimaku account and an API key (a personal access string) before this feature works. Create an account at [jimaku.cc](https://jimaku.cc), copy your key, and add it to your config as shown under [Configuration](#configuration) below. Without a key, the search modal will report "Jimaku API key not set."
@@ -14,21 +14,23 @@ When you open the modal, SubMiner parses the current video filename to extract a
 
 From there:
 
-1. **Search** - SubMiner queries the Jimaku API with the parsed title. Results appear as a list of anime entries (Japanese and English names).
-2. **Browse entries** - Select an entry to load its available subtitle files, filtered by episode if one was detected.
-3. **Browse files** - Files show name, size, and last-modified date. If a language preference is configured, files are sorted accordingly (e.g., Japanese-tagged files first).
-4. **Download** - Selecting a file downloads it to the same directory as the video (or a temp directory for remote/streamed media) and loads it into mpv as a new subtitle track.
+1. **Pick a catalogue** - The **Anime** and **Live action** tabs at the top of the modal choose which Jimaku catalogue to search. Switching tabs re-runs the current search. The choice persists until SubMiner restarts.
+2. **Search** - SubMiner queries the Jimaku API with the parsed title. Results appear as a list of entries (Japanese and English names).
+3. **Browse entries** - Select an entry to load its available subtitle files, filtered by episode if one was detected.
+4. **Browse files** - Files show name, size, and last-modified date. If a language preference is configured, files are sorted accordingly (e.g., Japanese-tagged files first).
+5. **Download** - Selecting a file downloads it to the same directory as the video (or a temp directory for remote/streamed media) and loads it into mpv as a new subtitle track.
 
 If no files match the current episode filter, a "Show all files" button lets you broaden the search to all episodes for that entry.
 
 ### Modal keyboard shortcuts
 
-| Key                       | Action                       |
-| ------------------------- | ---------------------------- |
-| `Enter` (in text field)   | Search                       |
-| `Enter` (in list)         | Select entry / download file |
-| `Arrow Up` / `Arrow Down` | Navigate entries or files    |
-| `Escape`                  | Close modal                  |
+| Key                          | Action                                        |
+| ---------------------------- | --------------------------------------------- |
+| `Enter` (in text field)      | Search                                        |
+| `Enter` (in list)            | Select entry / download file                  |
+| `Arrow Up` / `Arrow Down`    | Navigate entries or files                     |
+| `Arrow Left` / `Arrow Right` | Switch between the Anime and Live action tabs |
+| `Escape`                     | Close modal                                   |
 
 ## Configuration
 
@@ -52,7 +54,7 @@ Add a `jimaku` section to your `config.jsonc`:
 | `jimaku.apiKeyCommand`      | `string`                     | -                     | Shell command that prints the API key to stdout. Useful for secret managers (e.g., `pass jimaku/api-key`).                                               |
 | `jimaku.apiBaseUrl`         | `string`                     | `"https://jimaku.cc"` | Base URL for the Jimaku API. Only change this if using a mirror or local instance.                                                                       |
 | `jimaku.languagePreference` | `"ja"` \| `"en"` \| `"none"` | `"ja"`                | Sort subtitle files by language tag. `"ja"` pushes Japanese-tagged files to the top; `"en"` does the same for English. `"none"` preserves the API order. |
-| `jimaku.maxEntryResults`    | `number`                     | `10`                  | Maximum number of anime entries returned per search.                                                                                                     |
+| `jimaku.maxEntryResults`    | `number`                     | `10`                  | Maximum number of entries returned per search.                                                                                                           |
 
 The keyboard shortcut is configured separately under `shortcuts`:
 

docs-site/public/config.example.jsonc

@@ -547,7 +547,7 @@
   // ==========================================
   // AnkiConnect Integration
   // Automatic Anki updates and media generation options.
-  // Hot-reload: ankiConnect.ai.enabled, media.normalizeAudio/mirrorMpvVolume/reviewTiming, knownWords, nPlusOne, fields.word/audio/image/sentence/miscInfo, behavior.autoUpdateNewCards, isLapis.sentenceCardModel, isKiku.fieldGrouping, isSenren.fieldGrouping, and lapisKiku.wordCardKind update live while SubMiner is running.
+  // Hot-reload: ankiConnect.ai.enabled, media.normalizeAudio/mirrorMpvVolume/reviewTiming, knownWords, nPlusOne, fields.word/audio/wordAudio/image/sentence/miscInfo, behavior.autoUpdateNewCards, isLapis.sentenceCardModel, isKiku.fieldGrouping, isSenren.fieldGrouping, and lapisKiku.wordCardKind update live while SubMiner is running.
   // Shared AI provider transport settings are read from top-level ai and typically require restart.
   // Most other AnkiConnect settings still require restart.
   // ==========================================
@@ -568,6 +568,7 @@
     "fields": {
       "word": "Expression", // Card field for the mined word or expression text.
       "audio": "ExpressionAudio", // Card field that receives generated sentence audio.
+      "wordAudio": "ExpressionAudio", // Existing word-audio field read to time the frozen first frame of animated images. This mapping is only used for synchronization.
       "image": "Picture", // Card field that receives the captured screenshot or animated image.
       "sentence": "Sentence", // Card field that receives the source sentence text.
       "miscInfo": "MiscInfo", // Card field that receives the miscellaneous info pattern (see ankiConnect.metadata.pattern).
@@ -596,7 +597,7 @@
       "reviewTiming": false, // Review and preview subtitle media timing before SubMiner creates or enriches a mined card. Values: true | false
       "audioPadding": 0, // Seconds of padding appended to both ends of generated sentence audio and animated AVIF clips.
       "fallbackDuration": 3, // Fallback clip duration in seconds when subtitle timing data is unavailable.
-      "maxMediaDuration": 30 // Maximum allowed media clip duration in seconds.
+      "maxMediaDuration": 30 // Maximum allowed media clip duration in seconds. 0 disables the cap.
     }, // Media setting.
     "knownWords": {
       "highlightEnabled": false, // Enable fast local highlighting for words already known in Anki. Values: true | false
@@ -679,6 +680,16 @@
     "maxSearchResults": 10 // Maximum TsukiHime search results returned.
   }, // TsukiHime subtitle search configuration for Japanese primary and configured secondary subtitles. No API key required.
 
+  // ==========================================
+  // TMDB
+  // TMDB (The Movie Database) metadata for live-action dramas and movies in the stats Library: posters, synopses, and grouping by show.
+  // Hot-reload: TMDB changes apply to the next TMDB request.
+  // ==========================================
+  "tmdb": {
+    "apiKey": "", // Your own TMDB API key or read access token for live-action posters and synopses in the stats Library. Release builds bundle a project key, so set this only to use your own quota or when running from source (free under Settings > API on themoviedb.org).
+    "apiKeyCommand": "" // Shell command that prints the TMDB API key to stdout. Used instead of apiKey to avoid storing the key in plain text.
+  }, // TMDB (The Movie Database) metadata for live-action dramas and movies in the stats Library: posters, synopses, and grouping by show.
+
   // ==========================================
   // YouTube Playback Settings
   // Defaults for managed subtitle language preferences and YouTube subtitle loading.

docs-site/subtitle-generation.md

@@ -29,11 +29,21 @@ With speech detection configured, SubMiner keeps detected speech and other audib
 
 Passages that fit within Whisper's 30-second audio window stay intact. Longer passages prefer nearby detected speech starts when choosing cuts, falling back to quiet pauses, with a small overlap to provide context. This reduces early subtitles caused by starting a clip well before its dialogue, while keeping all retained audio covered. Matching overlapping cues are combined even when punctuation differs; repeated dialogue at separate times remains separate. SubMiner runs each passage in a fresh Whisper process so decoder state from earlier audio cannot affect later passages. This reloads the model for each passage and can increase generation time. Subtitle cues stay within the supplied audio and retain each passage's position on the original timeline. Progress reports the current passage.
 
-This mode favors retaining dialogue over excluding music, so songs and background sounds may also produce subtitles. It can take longer than transcribing only VAD-approved speech. Whisper can still miss or misrecognize dialogue, and its timestamps remain estimates. Uncheck **Focus on spoken dialogue** to return to full-audio transcription for the session, or clear `vadModelPath` to change the default. A selected detector or model that fails stops generation with an error. Existing subtitles are preserved.
+This mode favors retaining dialogue over excluding music, so songs and background sounds may also produce subtitles. It can take longer than transcribing only VAD-approved speech. Whisper can still miss or misrecognize dialogue, and its timestamps remain estimates. Uncheck **Focus on spoken dialogue** to disable VAD for the session, or clear `vadModelPath` to change the default. Without a usable subtitle reference, disabling VAD returns to full-audio transcription. A selected detector or model that fails stops generation with an error. Existing subtitles are preserved.
+
+## Using loaded subtitles as timing references
+
+When generating for the video currently open in mpv, SubMiner automatically looks for a dialogue subtitle track among its embedded subtitles and loaded external SRT, ASS/SSA, or WebVTT files. It prefers English, then tracks labeled full or dialogue. Forced tracks, image subtitles, generated subtitles, and tracks whose titles or filenames identify signs, songs, lyrics, karaoke, or opening/ending subtitles are skipped. These checks rely on metadata; an unlabeled signs-only file cannot always be identified.
+
+The selected reference appears in generation progress. SubMiner reads its timestamps, including the active primary or secondary track's subtitle delay, and uses nearby cue starts to guide cuts in long audio passages. Short passages stay intact. The reference works with or without **Focus on spoken dialogue**. With that option enabled, reference starts take priority over VAD starts when choosing a nearby cut; VAD still helps identify speech. Audio outside reference cues remains eligible for transcription, and Whisper still supplies the Japanese text and final timestamps. The reference is assumed to be timed for the playing video; this does not automatically sync a mistimed reference.
+
+Unreadable or empty references are skipped in favor of another eligible loaded track. If none can be read, generation uses its normal audio timing. The launcher uses loaded references only when its input matches the video currently open in mpv; standalone generation keeps its existing behavior. It captures reference tracks and delays together with the initial audio selection, before checking or downloading a model. When relying on mpv's selected audio, it stops and asks you to retry if the media changes or cannot be verified during capture.
 
 ## Choosing a model
 
-Start with **small** for a balance of Japanese recognition quality and CPU time. This is a general starting recommendation, not a benchmark for your hardware. Tiny and base need less memory and usually finish sooner, with more recognition errors. Medium and large models favor accuracy but need more resources. Large-v3-turbo is optimized for speed compared with large-v3, with some accuracy tradeoff; actual performance depends on your CPU, GPU, whisper.cpp build, and audio.
+The modal recommends **large-v3-turbo** when it detects an NVIDIA GPU through `nvidia-smi` and the selected `whisper-cli` discovers an available CUDA device. Otherwise it recommends **small** for a balance of Japanese recognition quality and CPU time. The check works before downloading a model and falls back to small if a tool is missing, fails, times out, or reports an unrecognized result. Vulkan, AMD, and Apple GPU support do not qualify for the turbo recommendation. Checks are cached for up to 30 seconds; changing the Whisper executable path triggers a new check.
+
+The recommendation labels the model picker and explains the detected support. It does not change your configured model, current selection, external model path, or launcher's model choice. It also does not force a GPU backend during transcription. These are starting recommendations, not hardware benchmarks or guarantees that every model fits in available GPU memory. Tiny and base need less memory and usually finish sooner, with more recognition errors. Medium and large models favor accuracy but need more resources. Large-v3-turbo is optimized for speed compared with large-v3, with some accuracy tradeoff; actual performance depends on your CPU, GPU, whisper.cpp build, and audio.
 
 The picker includes whisper.cpp's official multilingual tiny, base, small, medium, large-v1, large-v2, large-v3, and large-v3-turbo downloads, including their available quantized variants. Quantized models use less disk space and memory, with possible accuracy loss. English-only `.en` models are excluded. See the [upstream model list](https://github.com/ggml-org/whisper.cpp/blob/master/models/download-ggml-model.sh) and [Whisper's model guidance](https://github.com/openai/whisper#available-models-and-languages).
 
@@ -47,7 +57,7 @@ A configured external Model Path takes precedence and hides the managed model pi
 4. Optionally check **Focus on spoken dialogue** and click **Download speech detection model** if prompted.
 5. Click **Generate subtitles**.
 
-The modal shows audio preparation, transcription, and saving progress. Percentages appear when the underlying tool reports them. **Cancel** stops the current operation. Closing the modal lets the job continue; reopening it shows the current progress or result.
+The modal adapts to the player window, using a wider layout when space allows and scrolling in smaller windows. It shows audio preparation, transcription, and saving progress. Percentages appear when the underlying tool reports them. **Cancel** stops the current operation. Closing the modal lets the job continue; reopening it shows the current progress or result.
 
 **Escape** or **Close** closes the modal using the same focus and overlay restoration as other SubMiner modals. Change or disable its shortcut with `shortcuts.openSubtitleGeneration` in Settings. Ctrl+G remains assigned to field grouping.
 

docs-site/usage.md

@@ -93,6 +93,10 @@ delay. It also works with no running
 SubMiner app or mpv instance when you provide a file path. Omit the path to use the current
 mpv file and selected audio track.
 
+When the input matches the current mpv file, generation automatically uses an eligible embedded
+or loaded external subtitle track as a timing reference, preferring English dialogue and skipping
+tracks marked as signs, songs, or forced. See [timing references](/subtitle-generation#using-loaded-subtitles-as-timing-references).
+
 ```bash
 subminer generate-subs video.mkv --download-model
 subminer generate-subs video.mkv --model-path ~/models/ggml-medium.bin

docs-site/youtube-integration.md

@@ -149,6 +149,10 @@ These settings come from `config.jsonc` (or built-in defaults); there are no CLI
 - **Secondary subtitle fails**: Secondary track failures never block playback. The primary subtitle loads independently.
 - **Native mpv secondary rendering**: Stays hidden during YouTube flows so the SubMiner overlay remains the visible secondary subtitle surface.
 
+## Viewing stats
+
+The stats Library groups tracked YouTube videos by channel when channel metadata is available. Select **YouTube** in the Library filter to see those channels separately from anime. Each channel page lists its videos, watch time, vocabulary, and mined cards. See [Immersion Tracking](/immersion-tracking#library).
+
 ## Related pages
 
 - [Usage - YouTube Playback](/usage#youtube-playback)

docs/RELEASING.md

@@ -15,7 +15,10 @@
 
 Stable and prerelease workflows share `.github/workflows/package-release.yml`.
 Both callers explicitly pass the five required macOS signing/notarization
-secrets; `GITHUB_TOKEN` remains automatically available to the reusable workflow.
+secrets plus the optional `SUBMINER_TMDB_API_KEY` (the project TMDB key that
+`scripts/prepare-build-assets.mjs` stages into `dist/bundled-integration-keys.json`;
+artifacts built without it simply require users to set `tmdb.apiKey`).
+`GITHUB_TOKEN` remains automatically available to the reusable workflow.
 Each platform verifies its ASAR and external resources before signing, then
 measures the signed app and installers before upload. Missing runtime assets,
 foreign SQLite/Koffi binaries, duplicate UI fonts, demo media, source maps,

docs/architecture/README.md

@@ -30,8 +30,11 @@ Update checks and startup launcher migration share a serialized update-state sto
 
 - `src/main/` owns composition, runtime setup, IPC wiring, and app lifecycle adapters.
 - `src/main/boot/` owns boot-phase assembly seams so `src/main.ts` can stay focused on lifecycle coordination and startup-path selection.
+- `src/main/runtime/linux-overlay-mode-runtime.ts` owns Linux fullscreen mode state and window replacement. App cleanup cancels pending replacements; `main.ts` supplies window creation and subtitle refresh hooks.
 - `src/core/services/` owns focused runtime services plus pure or side-effect-bounded logic.
 - `src/core/services/subtitle-generation*.ts` shares local whisper.cpp transcription, safe model downloads, and progress between the launcher and Electron. Optional dialogue mode retains both Silero-detected speech and other audible sections, omits confidently silent gaps, decodes passages independently, and restores original media timing. `src/main/runtime/subtitle-generation-runtime.ts` owns the overlay job lifecycle and only loads completed subtitles into the same local media; `src/shared/subtitle-generation*.ts` owns configuration, the multilingual model catalog, and IPC contracts. The overlay runtime retains a session model selection, validates picker requests through IPC, and keeps external model paths authoritative.
+- Subtitle model recommendations use bounded `nvidia-smi` and Whisper CUDA discovery probes in `subtitle-generation-acceleration.ts`. The overlay runtime caches results by executable path for 30 seconds and exposes acceleration status through the existing status IPC. Recommendations do not alter model selection or transcription arguments.
+- `subtitle-generation-reference.ts` ranks mpv's loaded text subtitle tracks, excludes signs/songs and forced references, and extracts timing hints with FFmpeg. The overlay and launcher snapshot references only for matching media, including active subtitle delays. Hints guide long-passage cuts with or without VAD; they never limit audio coverage or replace Whisper timestamps.
 - `src/renderer/` owns overlay rendering and input behavior.
 - `src/config/` owns config definitions, defaults, loading, and resolution.
 - `src/types/` owns shared cross-runtime contracts via domain entrypoints; `src/types.ts` stays a compatibility barrel.

docs/architecture/domains.md

@@ -30,6 +30,7 @@ Read when: you need to find the owner module for a behavior or test surface
 - Immersion sync: `src/core/services/stats-sync/`, bound by `src/main/sync-cli.ts`.
   `snapshot-transfer.ts` selects compressed rsync or scp. `transfer-cache.ts` atomically retains the last successfully received snapshot per hashed peer/database identity under the config directory's `sync-transfer-cache/`. Cache copies seed isolated transfer directories; rsync verifies reconstructed files before the existing merge engine runs. The `--make-temp` / `--remove-temp` helpers accept an internal `--transfer-cache` key, with a fallback for older peers that do not recognize it.
 - AniList tracking + character dictionary: `src/core/services/anilist/`, `src/main/runtime/composers/anilist-*`, `src/main/character-dictionary-runtime.ts`, `src/main/character-dictionary-runtime/`
+- TMDB live-action metadata: `src/core/services/tmdb/` (client + exact-title resolver), `src/core/services/immersion-tracker/live-action-link.ts` (links an entry to a TMDB title and merges other holders of the same title). The AniList cover-art fetcher calls the resolver as its fallback; `imm_anime.media_kind` marks the result and keeps the entry out of AniList season repair.
 - Jellyfin integration: `src/core/services/jellyfin*.ts`, `src/main/runtime/composers/jellyfin-*`
 - Anime browser: extension bridge client, sidecar, and stream handling in `src/anime-bridge/`;
   the loopback stream proxy separates request transport/retry from response transformation in

docs/workflow/verification.md

@@ -22,9 +22,15 @@ Read when: selecting the right verification lane for a change
   pull requests, stable tags, and prerelease tags. Keep common quality steps
   there instead of copying them into caller workflows.
 - The reusable gate installs Lua and runs `bun run test:env`, so the shipped mpv
-  plugin tests run for every pull request and tagged release.
+  plugin tests and launcher smoke run for every pull request and tagged release.
   Lua installation uses only the runner's Ubuntu package sources so unrelated
   third-party repository failures do not block the gate.
+- In the reusable gate, `test:coverage:src` is also the blocking execution of the
+  discovered `src/**` test lane. The coverage runner returns the failing test's
+  status, so CI does not rerun that lane through `test:fast`. Launcher unit and
+  script tests still run separately because they are outside the coverage lane.
+- Launcher smoke artifacts are uploaded after `test:env` fails. CI does not rerun
+  launcher smoke solely to collect the same artifacts.
 
 ## Default Handoff Gate
 
@@ -49,7 +55,7 @@ bun run docs:build
 - Internal KB, `AGENTS.md`, or `.agents/skills/**` changes: `bun run test:docs:kb`
 - Config/schema/defaults: `bun run test:config`, then `bun run generate:config-example` if template/defaults changed
 - Launcher/plugin: `bun run test:launcher` or `bun run test:env`
-- Runtime-compat / compiled behavior: `bun run test:runtime:compat`
+- Runtime-compat / compiled behavior after `bun run build`: `bun run test:runtime:compat`
 - Stats dashboard UI: `bun run test:stats`
 - Build/release scripts (`scripts/**`): `bun run test:scripts`
 - Packaging: build the platform package, then run `bun run test:package <resources-directory>`.
@@ -62,11 +68,30 @@ bun run docs:build
 
 ## Coverage Reporting
 
-- `bun run test:coverage:src` runs the maintained `test:src` lane through a sharded coverage runner: one Bun coverage process per test file, then merged LCOV output.
+- `bun run test:coverage:src` runs the same discovered `bun-src-full` membership as
+  `test:src` through a sharded coverage runner: one Bun coverage process per test
+  file, then merged LCOV output.
+- A failing coverage shard stops the runner with a nonzero status. Coverage is a
+  source test gate, not a report-only step.
 - Machine-readable output lands at `coverage/test-src/lcov.info`.
 - Every reusable quality-gate run uploads that LCOV file as the
   `coverage-test-src` artifact.
 
+## Compiled Runtime Smoke
+
+- `bun run test:smoke:dist` and its `test:runtime:compat` alias require an existing
+  full build and fail with the missing artifact paths when `dist/` or the stats UI
+  bundle is absent.
+- The check runs the emitted stats daemon under Electron's Node runtime. It opens
+  the production HTTP server, queries the overview endpoint through native
+  libsql-backed storage, and leaves an HTTP request body unfinished before
+  shutting the daemon down. It verifies a clean exit and that the port and
+  ownership state are released despite the unfinished request.
+- The check also occupies the configured port, requires startup to fail without
+  stale ownership state, releases the conflict, and verifies a clean retry.
+- This is not a full Electron UI startup check. It does not require a display and
+  makes no claims about renderer, window, tray, or mpv behavior.
+
 ## Dependency Audit Policy
 
 - `bun audit --audit-level high` blocks the reusable quality gate.

launcher/commands/generate-subtitles-command.test.ts

@@ -136,25 +136,28 @@ test('current mpv generation requires an identifiable selected audio track', asy
   }
 });
 
-test('explicit local file leaves Japanese track selection to the shared generator', async () => {
+test('explicit playing file leaves audio selection to the generator but inspects subtitle references', async () => {
   const f = fixture();
   await runGenerateSubtitlesCommand(f.context, f.deps);
   assert.equal(f.generations[0]?.audioStreamIndex, undefined);
   assert.equal(
     f.commands.some((command) => command[1] === 'track-list'),
-    false,
+    true,
   );
 });
 
 test('launcher does not load generated subtitles after mpv switches files', async () => {
   const f = fixture(['generate-subs']);
-  let pathRequests = 0;
+  let changed = false;
   f.deps.mpvCommand = async (_socket, command) => {
     f.commands.push(command);
-    if (command[1] === 'path')
-      return ++pathRequests === 1 ? '/media/episode.mkv' : '/media/next.mkv';
+    if (command[1] === 'path') return changed ? '/media/next.mkv' : '/media/episode.mkv';
     return [{ type: 'audio', selected: true, 'ff-index': 2 }];
   };
+  f.deps.generate = async () => {
+    changed = true;
+    return '/media/episode.ja.srt';
+  };
   await runGenerateSubtitlesCommand(f.context, f.deps);
   assert.equal(
     f.commands.some((command) => command[0] === 'sub-add'),
@@ -163,6 +166,86 @@ test('launcher does not load generated subtitles after mpv switches files', asyn
   assert.match(f.output.join(''), /Saved Japanese subtitles/);
 });
 
+test('current mpv generation rejects tracks when playback changes during capture', async () => {
+  for (const nextMedia of ['/media/next.mkv', null]) {
+    const f = fixture(['generate-subs']);
+    let media: string | null = '/media/episode.mkv';
+    let modelChecks = 0;
+    f.deps.mpvCommand = async (_socket, command) => {
+      if (command[1] === 'path') return media;
+      if (command[1] === 'track-list') {
+        media = nextMedia;
+        return [{ type: 'audio', selected: true, 'ff-index': 99 }];
+      }
+      return undefined;
+    };
+    f.deps.resolveModel = async () => {
+      modelChecks += 1;
+      return { kind: 'external', path: '/models/model.bin' };
+    };
+    await assert.rejects(
+      runGenerateSubtitlesCommand(f.context, f.deps),
+      /mpv media.*Run generate-subs again/,
+    );
+    assert.equal(f.generations.length, 0);
+    assert.equal(modelChecks, 0);
+  }
+});
+
+test('explicit media or audio stream remains usable when the mpv snapshot changes', async () => {
+  for (const options of [['/media/episode.mkv'], ['--audio-stream', '7']]) {
+    const f = fixture(['generate-subs', ...options]);
+    let media = '/media/episode.mkv';
+    f.deps.mpvCommand = async (_socket, command) => {
+      if (command[1] === 'path') return media;
+      if (command[1] === 'track-list') {
+        media = '/media/next.mkv';
+        return [
+          { type: 'audio', selected: true, 'ff-index': 99 },
+          { type: 'sub', lang: 'eng', 'ff-index': 100 },
+        ];
+      }
+      return undefined;
+    };
+    await runGenerateSubtitlesCommand(f.context, f.deps);
+    assert.equal(f.generations[0]?.mediaPath, '/media/episode.mkv');
+    assert.equal(
+      f.generations[0]?.audioStreamIndex,
+      options[0] === '--audio-stream' ? 7 : undefined,
+    );
+    assert.deepEqual(f.generations[0]?.references, []);
+  }
+});
+
+test('launcher captures loaded external references only for the matching media', async () => {
+  for (const matching of [true, false]) {
+    const f = fixture();
+    f.deps.mpvCommand = async (_socket, command) => {
+      if (command[1] === 'path') return matching ? '/media/episode.mkv' : '/media/other.mkv';
+      if (command[1] === 'working-directory') return '/mpv';
+      if (command[1] === 'track-list')
+        return [
+          { type: 'sub', external: true, 'external-filename': 'episode.en.signs.ass' },
+          { type: 'sub', external: true, 'external-filename': 'episode.en.srt' },
+        ];
+      return undefined;
+    };
+    await runGenerateSubtitlesCommand(f.context, f.deps);
+    assert.deepEqual(
+      f.generations[0]?.references,
+      matching
+        ? [
+            {
+              label: 'episode.en.srt',
+              delaySeconds: 0,
+              source: { kind: 'external', path: '/mpv/episode.en.srt' },
+            },
+          ]
+        : [],
+    );
+  }
+});
+
 test('explicit managed model overrides external config and downloads before generation', async () => {
   const f = fixture([
     'generate-subs',
@@ -191,6 +274,69 @@ test('explicit managed model overrides external config and downloads before gene
   assert.equal(f.generations.length, 1);
 });
 
+test('audio and subtitle timing use the initial mpv snapshot across model setup', async () => {
+  for (const changeDuring of ['resolve', 'download']) {
+    const f = fixture(['generate-subs', '--download-model']);
+    let changed = false;
+    let trackReads = 0;
+    f.deps.mpvCommand = async (_socket, command) => {
+      if (command[1] === 'path') return '/media/episode.mkv';
+      if (command[1] === 'track-list') {
+        trackReads += 1;
+        return [
+          { type: 'audio', selected: true, 'ff-index': changed ? 3 : 2 },
+          { type: 'sub', id: 1, title: 'English Full', lang: 'eng', 'ff-index': changed ? 5 : 4 },
+        ];
+      }
+      if (command[1] === 'sid') return 1;
+      if (command[1] === 'sub-delay') return changed ? 9 : 1.5;
+      return undefined;
+    };
+    f.deps.resolveModel = async () => {
+      if (changeDuring === 'resolve') changed = true;
+      return { kind: 'missing', path: '/models/model.bin' };
+    };
+    f.deps.downloadModel = async () => {
+      changed = true;
+      return '/models/model.bin';
+    };
+    await runGenerateSubtitlesCommand(f.context, f.deps);
+    assert.equal(f.generations[0]?.audioStreamIndex, 2);
+    assert.deepEqual(f.generations[0]?.references, [
+      {
+        label: 'English Full',
+        delaySeconds: 1.5,
+        source: { kind: 'embedded', streamIndex: 4 },
+      },
+    ]);
+    assert.equal(trackReads, 1);
+  }
+});
+
+test('explicit audio stream bypasses mpv audio selection while retaining subtitle references', async () => {
+  const f = fixture(['generate-subs', '--audio-stream', '7']);
+  f.deps.mpvCommand = async (_socket, command) => {
+    if (command[1] === 'path') return '/media/episode.mkv';
+    if (command[1] === 'track-list')
+      return [
+        { type: 'audio', selected: true, external: true, 'ff-index': 0 },
+        { type: 'sub', id: 2, title: 'English Full', lang: 'eng', 'ff-index': 4 },
+      ];
+    if (command[1] === 'secondary-sid') return 2;
+    if (command[1] === 'secondary-sub-delay') return -0.5;
+    return undefined;
+  };
+  await runGenerateSubtitlesCommand(f.context, f.deps);
+  assert.equal(f.generations[0]?.audioStreamIndex, 7);
+  assert.deepEqual(f.generations[0]?.references, [
+    {
+      label: 'English Full',
+      delaySeconds: -0.5,
+      source: { kind: 'embedded', streamIndex: 4 },
+    },
+  ]);
+});
+
 test('generation can run standalone and never loads subtitles into another video', async () => {
   for (const playing of [null, '/media/different.mkv']) {
     const f = fixture();

launcher/commands/generate-subtitles-command.ts

@@ -8,6 +8,10 @@ import {
   resolveSubtitleGenerationTools,
 } from '../../src/core/services/subtitle-generation.js';
 import { requireSubtitleGenerationTools } from '../../src/core/services/subtitle-generation-tools.js';
+import {
+  readSubtitleGenerationReferences,
+  type SubtitleGenerationReference,
+} from '../../src/core/services/subtitle-generation-reference.js';
 import {
   resolveSubtitleGenerationConfig,
   type SubtitleGenerationProgress,
@@ -69,12 +73,9 @@ async function readMpvMedia(socketPath: string, command: GenerationCommandDeps['
   return localMediaPath(media, workingDirectory);
 }
 
-async function readMpvAudioStream(
-  socketPath: string,
-  command: GenerationCommandDeps['mpvCommand'],
-) {
-  const tracks = await command(socketPath, ['get_property', 'track-list'], 1000);
-  for (const track of Array.isArray(tracks) ? tracks : []) {
+function selectedMpvAudioStream(value: unknown) {
+  const tracks: unknown[] = Array.isArray(value) ? value : [];
+  for (const track of tracks) {
     if (
       typeof track === 'object' &&
       track !== null &&
@@ -160,11 +161,29 @@ export async function runGenerateSubtitlesCommand(
     const mediaPath = options.mediaPath ? localMediaPath(options.mediaPath) : currentMedia;
     if (!mediaPath)
       throw new Error('Pass a local video file or open one in mpv before running generate-subs.');
+    // Capture audio and subtitle timing together before model setup can yield to playback changes.
+    const matchesCurrentMedia = currentMedia !== null && sameFile(currentMedia, mediaPath);
+    const tracks = matchesCurrentMedia
+      ? await deps
+          .mpvCommand(context.mpvSocketPath, ['get_property', 'track-list'], 1000)
+          .catch(() => null)
+      : null;
+    let references: SubtitleGenerationReference[] = [];
+    if (matchesCurrentMedia) {
+      const candidates = await readSubtitleGenerationReferences(tracks, (name) =>
+        deps.mpvCommand(context.mpvSocketPath, ['get_property', name], 1000),
+      );
+      const stillPlaying = await readMpvMedia(context.mpvSocketPath, deps.mpvCommand).catch(
+        () => null,
+      );
+      if (stillPlaying && sameFile(stillPlaying, mediaPath)) references = candidates;
+      else if (!options.mediaPath && options.audioStreamIndex === undefined)
+        throw new Error(
+          'The current mpv media changed or could not be verified while reading tracks. Run generate-subs again.',
+        );
+    }
     const audioStreamIndex =
-      options.audioStreamIndex ??
-      (!options.mediaPath
-        ? await readMpvAudioStream(context.mpvSocketPath, deps.mpvCommand)
-        : undefined);
+      options.audioStreamIndex ?? (!options.mediaPath ? selectedMpvAudioStream(tracks) : undefined);
     const onProgress = createGenerationProgressReporter(write);
     // Missing executables fail here, before any model download starts.
     requireSubtitleGenerationTools(await deps.resolveTools(config));
@@ -183,6 +202,7 @@ export async function runGenerateSubtitlesCommand(
       modelDirectory,
       mediaPath,
       audioStreamIndex,
+      references,
       outputPath: options.outputPath
         ? path.resolve(resolvePathMaybe(options.outputPath))
         : undefined,

launcher/test-support/immersion-db-schema.test.ts

@@ -31,6 +31,7 @@ const SYNC_SCHEMA_OBJECTS = [
   'imm_lifetime_applied_sessions',
   'imm_stats_excluded_words',
   'idx_anime_normalized_title',
+  'idx_anime_namespace_title',
   'idx_anime_anilist_id',
   'idx_videos_anime_id',
   'idx_sessions_video_started',

launcher/test-support/immersion-db-schema.ts

@@ -12,7 +12,7 @@ export const IMMERSION_DB_FIXTURE_DDL = `
   );
   CREATE TABLE imm_anime(
     anime_id INTEGER PRIMARY KEY AUTOINCREMENT,
-    normalized_title_key TEXT NOT NULL UNIQUE,
+    normalized_title_key TEXT NOT NULL,
     canonical_title TEXT NOT NULL,
     anilist_id INTEGER UNIQUE,
     title_romaji TEXT,
@@ -20,10 +20,14 @@ export const IMMERSION_DB_FIXTURE_DDL = `
     title_native TEXT,
     episodes_total INTEGER,
     description TEXT,
+    media_kind TEXT NOT NULL DEFAULT 'anime',
+    tmdb_id INTEGER,
+    tmdb_type TEXT,
     metadata_json TEXT,
     CREATED_DATE TEXT,
     LAST_UPDATE_DATE TEXT
   );
+  CREATE UNIQUE INDEX idx_anime_namespace_title ON imm_anime((media_kind = 'youtube'), normalized_title_key);
   CREATE TABLE imm_videos(
     video_id INTEGER PRIMARY KEY AUTOINCREMENT,
     video_key TEXT NOT NULL UNIQUE,

package.json

@@ -52,7 +52,7 @@
     "test:docs:kb": "bun test scripts/docs-knowledge-base.test.ts",
     "test:plugin:src": "lua scripts/test-plugin-lua-compat.lua && lua scripts/test-plugin-start-gate.lua && lua scripts/test-plugin-process-start-retries.lua && lua scripts/test-plugin-restart-feedback.lua && lua scripts/test-plugin-session-bindings.lua && lua scripts/test-plugin-binary-windows.lua",
     "test:launcher:smoke:src": "bun test launcher/smoke.e2e.test.ts",
-    "test:smoke:dist": "bun scripts/run-test-lane.mjs bun-src-full",
+    "test:smoke:dist": "env ELECTRON_RUN_AS_NODE=1 electron scripts/compiled-runtime-smoke.mjs",
     "test:subtitle:src": "bun test src/core/services/subsync.test.ts src/subsync/utils.test.ts",
     "test:immersion:sqlite:src": "bun test src/core/services/immersion-tracker-service.test.ts src/core/services/immersion-tracker/storage-session.test.ts",
     "test:immersion:sqlite:dist": "bun test dist/core/services/immersion-tracker-service.test.js dist/core/services/immersion-tracker/storage-session.test.js",
@@ -64,7 +64,7 @@
     "test:scripts": "bun scripts/run-test-lane.mjs scripts",
     "test:stats": "bun scripts/run-test-lane.mjs stats",
     "test:env": "bun run test:launcher:smoke:src && bun run test:plugin:src && bun run test:immersion:sqlite:src",
-    "test:runtime:compat": "bun run tsc && bun scripts/run-test-lane.mjs bun-src-full",
+    "test:runtime:compat": "bun run test:smoke:dist",
     "test": "bun run test:fast",
     "test:config": "bun scripts/run-test-lane.mjs config",
     "test:launcher": "bun scripts/run-test-lane.mjs launcher && bun run test:plugin:src",

scripts/bundled-integration-keys.mjs

@@ -0,0 +1,29 @@
+import fs from 'node:fs';
+import path from 'node:path';
+
+/**
+ * Release builds carry a project-owned TMDB key so live-action lookups work
+ * without user setup. The key is injected from the SUBMINER_TMDB_API_KEY
+ * environment variable at build time (a GitHub Actions secret in CI) and never
+ * lives in the repository. The runtime reader is
+ * src/core/services/tmdb/bundled-api-key.ts; keep the file name in sync.
+ */
+export const BUNDLED_INTEGRATION_KEYS_FILENAME = 'bundled-integration-keys.json';
+export const TMDB_API_KEY_ENV = 'SUBMINER_TMDB_API_KEY';
+
+/**
+ * Write the bundled keys file into `distDir`, or remove a stale one when no
+ * key is present so a keyless build never ships an older key by accident.
+ * Returns the names of the keys staged.
+ */
+export function stageBundledIntegrationKeys(distDir, env = process.env) {
+  const outputPath = path.join(distDir, BUNDLED_INTEGRATION_KEYS_FILENAME);
+  const tmdbApiKey = env[TMDB_API_KEY_ENV]?.trim() ?? '';
+  if (!tmdbApiKey) {
+    fs.rmSync(outputPath, { force: true });
+    return [];
+  }
+  fs.mkdirSync(distDir, { recursive: true });
+  fs.writeFileSync(outputPath, `${JSON.stringify({ tmdbApiKey })}\n`, { mode: 0o644 });
+  return ['tmdb'];
+}

scripts/bundled-integration-keys.test.ts

@@ -0,0 +1,39 @@
+import assert from 'node:assert/strict';
+import fs from 'node:fs';
+import os from 'node:os';
+import path from 'node:path';
+import test from 'node:test';
+import {
+  BUNDLED_INTEGRATION_KEYS_FILENAME,
+  stageBundledIntegrationKeys,
+} from './bundled-integration-keys.mjs';
+
+function withDistDir(work: (distDir: string) => void): void {
+  const distDir = fs.mkdtempSync(path.join(os.tmpdir(), 'subminer-bundled-keys-'));
+  try {
+    work(distDir);
+  } finally {
+    fs.rmSync(distDir, { recursive: true, force: true });
+  }
+}
+
+test('stageBundledIntegrationKeys writes the TMDB key from the environment', () => {
+  withDistDir((distDir) => {
+    const staged = stageBundledIntegrationKeys(distDir, { SUBMINER_TMDB_API_KEY: ' abc123 ' });
+    assert.deepEqual(staged, ['tmdb']);
+    const written = JSON.parse(
+      fs.readFileSync(path.join(distDir, BUNDLED_INTEGRATION_KEYS_FILENAME), 'utf8'),
+    );
+    assert.deepEqual(written, { tmdbApiKey: 'abc123' });
+  });
+});
+
+test('stageBundledIntegrationKeys removes a stale file when the variable is unset', () => {
+  withDistDir((distDir) => {
+    const outputPath = path.join(distDir, BUNDLED_INTEGRATION_KEYS_FILENAME);
+    fs.writeFileSync(outputPath, '{"tmdbApiKey":"old"}');
+    assert.deepEqual(stageBundledIntegrationKeys(distDir, {}), []);
+    assert.equal(fs.existsSync(outputPath), false);
+    assert.deepEqual(stageBundledIntegrationKeys(distDir, { SUBMINER_TMDB_API_KEY: '  ' }), []);
+  });
+});

scripts/compiled-runtime-smoke.mjs

@@ -0,0 +1,345 @@
+import assert from 'node:assert/strict';
+import { spawn } from 'node:child_process';
+import { existsSync, mkdtempSync, readFileSync, rmSync, statSync, writeFileSync } from 'node:fs';
+import { createServer } from 'node:net';
+import { request as httpRequest } from 'node:http';
+import { tmpdir } from 'node:os';
+import { dirname, join, resolve } from 'node:path';
+import { fileURLToPath } from 'node:url';
+
+const START_TIMEOUT_MS = 15_000;
+const STOP_TIMEOUT_MS = 5_000;
+const POLL_INTERVAL_MS = 25;
+const MAX_CHILD_OUTPUT_BYTES = 64 * 1024;
+
+const scriptDir = dirname(fileURLToPath(import.meta.url));
+const repoRootArgIndex = process.argv.indexOf('--repo-root');
+const repoRoot = resolve(
+  repoRootArgIndex === -1 ? join(scriptDir, '..') : (process.argv[repoRootArgIndex + 1] ?? ''),
+);
+const paths = {
+  mainEntry: join(repoRoot, 'dist', 'main-entry.js'),
+  daemonEntry: join(repoRoot, 'dist', 'stats-daemon-runner.js'),
+  statsServer: join(repoRoot, 'dist', 'core', 'services', 'stats-server.js'),
+  tracker: join(repoRoot, 'dist', 'core', 'services', 'immersion-tracker-service.js'),
+  statsIndex: join(repoRoot, 'stats', 'dist', 'index.html'),
+};
+
+function requireCompiledArtifacts() {
+  const missing = Object.values(paths).filter((artifactPath) => !existsSync(artifactPath));
+  if (missing.length > 0) {
+    throw new Error(
+      `Compiled runtime artifacts are missing. Run \`bun run build\` before this check:\n${missing
+        .map((artifactPath) => `  - ${artifactPath}`)
+        .join('\n')}`,
+    );
+  }
+}
+
+function requireElectronNodeRuntime() {
+  if (!process.versions.electron || process.env.ELECTRON_RUN_AS_NODE !== '1') {
+    throw new Error(
+      'This check must run with Electron in Node mode. Use `bun run test:smoke:dist`.',
+    );
+  }
+}
+
+function delay(ms) {
+  return new Promise((resolvePromise) => setTimeout(resolvePromise, ms));
+}
+
+function listen(server, port = 0) {
+  return new Promise((resolvePromise, reject) => {
+    const onError = (error) => {
+      server.off('listening', onListening);
+      reject(error);
+    };
+    const onListening = () => {
+      server.off('error', onError);
+      const address = server.address();
+      if (!address || typeof address === 'string') {
+        reject(new Error('Could not resolve the stats smoke port.'));
+        return;
+      }
+      resolvePromise(address.port);
+    };
+    server.once('error', onError);
+    server.once('listening', onListening);
+    server.listen(port, '127.0.0.1');
+  });
+}
+
+function closeServer(server) {
+  return new Promise((resolvePromise, reject) => {
+    server.close((error) => {
+      if (error) {
+        reject(error);
+        return;
+      }
+      resolvePromise();
+    });
+  });
+}
+
+async function findAvailablePort() {
+  const reservation = createServer();
+  const port = await listen(reservation);
+  await closeServer(reservation);
+  return port;
+}
+
+function writeSmokeConfig(userDataPath, port) {
+  writeFileSync(
+    join(userDataPath, 'config.json'),
+    `${JSON.stringify({ stats: { serverPort: port } })}\n`,
+  );
+}
+
+function spawnDaemon(userDataPath, responsePath) {
+  const child = spawn(
+    process.execPath,
+    [
+      paths.daemonEntry,
+      '--stats-user-data-path',
+      userDataPath,
+      '--stats-response-path',
+      responsePath,
+    ],
+    {
+      cwd: repoRoot,
+      env: { ...process.env, ELECTRON_RUN_AS_NODE: '1' },
+      stdio: ['ignore', 'pipe', 'pipe'],
+    },
+  );
+  let output = '';
+  const appendOutput = (chunk) => {
+    if (output.length >= MAX_CHILD_OUTPUT_BYTES) return;
+    output += String(chunk);
+    if (output.length > MAX_CHILD_OUTPUT_BYTES) {
+      output = `${output.slice(0, MAX_CHILD_OUTPUT_BYTES)}\n[child output truncated]\n`;
+    }
+  };
+  child.stdout.on('data', appendOutput);
+  child.stderr.on('data', appendOutput);
+  let exited = false;
+  const exit = new Promise((resolvePromise) => {
+    child.once('error', (error) => {
+      exited = true;
+      resolvePromise({ code: null, signal: null, error });
+    });
+    child.once('exit', (code, signal) => {
+      exited = true;
+      resolvePromise({ code, signal, error: null });
+    });
+  });
+  return { child, exit, hasExited: () => exited, getOutput: () => output };
+}
+
+async function waitForResponse(responsePath, daemon) {
+  const deadline = Date.now() + START_TIMEOUT_MS;
+  while (Date.now() < deadline) {
+    if (existsSync(responsePath)) {
+      try {
+        return JSON.parse(readFileSync(responsePath, 'utf8'));
+      } catch {
+        // The daemon may still be finishing the response file write.
+      }
+    }
+    if (daemon.hasExited()) {
+      const result = await daemon.exit;
+      throw new Error(
+        `Stats daemon exited before writing a startup response (${formatExit(result)}).\n${daemon.getOutput()}`,
+      );
+    }
+    await delay(POLL_INTERVAL_MS);
+  }
+  throw new Error(`Timed out waiting for stats daemon startup.\n${daemon.getOutput()}`);
+}
+
+function formatExit(result) {
+  if (result.error) return result.error.message;
+  if (result.signal) return `signal ${result.signal}`;
+  return `exit code ${result.code}`;
+}
+
+async function waitForExit(daemon, timeoutMs) {
+  return await Promise.race([daemon.exit, delay(timeoutMs).then(() => null)]);
+}
+
+async function stopDaemon(daemon) {
+  if (daemon.hasExited()) {
+    return await daemon.exit;
+  }
+  daemon.child.kill('SIGTERM');
+  const result = await waitForExit(daemon, STOP_TIMEOUT_MS);
+  if (result) return result;
+  daemon.child.kill('SIGKILL');
+  await daemon.exit;
+  throw new Error(`Stats daemon did not stop after SIGTERM.\n${daemon.getOutput()}`);
+}
+
+async function assertPortCanBind(port) {
+  const server = createServer();
+  try {
+    await listen(server, port);
+  } finally {
+    if (server.listening) await closeServer(server);
+  }
+}
+
+async function fetchOverview(url, daemon) {
+  const deadline = Date.now() + START_TIMEOUT_MS;
+  let lastError = null;
+  while (Date.now() < deadline) {
+    try {
+      return await fetch(`${url}/api/stats/overview`, {
+        signal: AbortSignal.timeout(START_TIMEOUT_MS),
+      });
+    } catch (error) {
+      lastError = error;
+    }
+    if (daemon.hasExited()) {
+      const result = await daemon.exit;
+      throw new Error(
+        `Stats daemon exited before accepting HTTP requests (${formatExit(result)}).\n${daemon.getOutput()}`,
+      );
+    }
+    await delay(POLL_INTERVAL_MS);
+  }
+  throw new Error(
+    `Timed out waiting for the stats HTTP server: ${lastError instanceof Error ? lastError.message : String(lastError)}\n${daemon.getOutput()}`,
+  );
+}
+
+async function runHealthyStartup(userDataPath, port, responseName) {
+  writeSmokeConfig(userDataPath, port);
+  const responsePath = join(userDataPath, responseName);
+  const statePath = join(userDataPath, 'stats-daemon.json');
+  const databasePath = join(userDataPath, 'immersion.sqlite');
+  const daemon = spawnDaemon(userDataPath, responsePath);
+  let shutdownResult;
+  let unfinishedRequest;
+
+  try {
+    const startup = await waitForResponse(responsePath, daemon);
+    assert.deepEqual(startup, { ok: true, url: `http://127.0.0.1:${port}` });
+
+    const response = await fetchOverview(startup.url, daemon);
+    assert.equal(response.status, 200);
+    assert.match(response.headers.get('content-type') ?? '', /^application\/json\b/);
+    const overview = await response.json();
+    assert.equal(typeof overview, 'object');
+    assert.ok(overview !== null);
+    assert.ok(Array.isArray(overview.sessions));
+    assert.ok(Array.isArray(overview.rollups));
+    assert.equal(typeof overview.hints, 'object');
+
+    const dashboard = await fetch(`${startup.url}/?overlay=1`, {
+      signal: AbortSignal.timeout(START_TIMEOUT_MS),
+    });
+    assert.equal(dashboard.status, 200);
+    assert.match(dashboard.headers.get('content-type') ?? '', /^text\/html\b/);
+    assert.match(await dashboard.text(), /id="root"/);
+
+    assert.ok(existsSync(databasePath), 'The compiled tracker did not create its SQLite database.');
+    assert.ok(
+      statSync(databasePath).size > 0,
+      'The compiled tracker created an empty SQLite file.',
+    );
+
+    // Leave a real request body unfinished so shutdown must bound its drain wait.
+    unfinishedRequest = httpRequest(new URL('/api/stats/anki/notesInfo', startup.url), {
+      method: 'POST',
+      headers: {
+        'Content-Type': 'application/json',
+        'Content-Length': '1000',
+        Expect: '100-continue',
+      },
+      signal: AbortSignal.timeout(START_TIMEOUT_MS),
+    });
+    await new Promise((resolvePromise, reject) => {
+      unfinishedRequest.on('error', reject);
+      unfinishedRequest.once('continue', () => {
+        unfinishedRequest.write('{');
+        resolvePromise();
+      });
+      unfinishedRequest.flushHeaders();
+    });
+  } finally {
+    try {
+      shutdownResult = await stopDaemon(daemon);
+    } finally {
+      unfinishedRequest?.destroy();
+    }
+  }
+
+  assert.equal(shutdownResult.code, 0, `Stats daemon shutdown failed.\n${daemon.getOutput()}`);
+  assert.equal(existsSync(statePath), false, 'Stats daemon state remained after shutdown.');
+  await assertPortCanBind(port);
+}
+
+async function runConflictRecovery() {
+  const userDataPath = mkdtempSync(join(tmpdir(), 'subminer-compiled-conflict-'));
+  const reservation = createServer();
+  const port = await listen(reservation);
+  const responsePath = join(userDataPath, 'conflict-response.json');
+  const statePath = join(userDataPath, 'stats-daemon.json');
+  writeSmokeConfig(userDataPath, port);
+  const daemon = spawnDaemon(userDataPath, responsePath);
+  const failures = [];
+
+  try {
+    const response = await waitForResponse(responsePath, daemon);
+    if (response.ok !== false) {
+      failures.push('The stats daemon reported success while its configured port was occupied.');
+    }
+    const result = await waitForExit(daemon, STOP_TIMEOUT_MS);
+    if (!result) {
+      failures.push('The stats daemon did not exit after its configured port failed to bind.');
+    } else if (result.code === 0) {
+      failures.push(
+        'The stats daemon exited successfully after its configured port failed to bind.',
+      );
+    }
+    if (existsSync(statePath)) {
+      failures.push(
+        'The stats daemon left ownership state behind after its configured port failed to bind.',
+      );
+    }
+  } finally {
+    await closeServer(reservation);
+    if (!daemon.hasExited()) {
+      await stopDaemon(daemon);
+    }
+  }
+
+  try {
+    await runHealthyStartup(userDataPath, port, 'recovery-response.json');
+  } finally {
+    rmSync(userDataPath, { recursive: true, force: true });
+  }
+
+  assert.deepEqual(failures, []);
+}
+
+async function main() {
+  requireCompiledArtifacts();
+  requireElectronNodeRuntime();
+
+  const userDataPath = mkdtempSync(join(tmpdir(), 'subminer-compiled-runtime-'));
+  try {
+    await runHealthyStartup(userDataPath, await findAvailablePort(), 'startup-response.json');
+  } finally {
+    rmSync(userDataPath, { recursive: true, force: true });
+  }
+  await runConflictRecovery();
+
+  process.stdout.write(
+    `Compiled runtime smoke passed with Electron ${process.versions.electron}, Node ${process.versions.node}, HTTP, and native SQLite.\n`,
+  );
+}
+
+main().catch((error) => {
+  process.stderr.write(`${error instanceof Error ? error.stack : String(error)}\n`);
+  process.exitCode = 1;
+});

scripts/compiled-runtime-smoke.test.ts

@@ -0,0 +1,28 @@
+import assert from 'node:assert/strict';
+import { spawnSync } from 'node:child_process';
+import fs from 'node:fs';
+import os from 'node:os';
+import path from 'node:path';
+import test from 'node:test';
+
+test('compiled runtime smoke fails clearly when build artifacts are missing', () => {
+  const emptyRepo = fs.mkdtempSync(path.join(os.tmpdir(), 'subminer-compiled-missing-'));
+  try {
+    const result = spawnSync(
+      process.execPath,
+      ['scripts/compiled-runtime-smoke.mjs', '--repo-root', emptyRepo],
+      {
+        cwd: path.resolve(import.meta.dir, '..'),
+        encoding: 'utf8',
+      },
+    );
+
+    assert.equal(result.status, 1);
+    assert.match(result.stderr, /Compiled runtime artifacts are missing/);
+    assert.match(result.stderr, /dist\/main-entry\.js/);
+    assert.match(result.stderr, /dist\/stats-daemon-runner\.js/);
+    assert.doesNotMatch(result.stderr, /must run with Electron/);
+  } finally {
+    fs.rmSync(emptyRepo, { recursive: true, force: true });
+  }
+});

scripts/prepare-build-assets.mjs

@@ -3,6 +3,7 @@ import os from 'node:os';
 import path from 'node:path';
 import { execFileSync } from 'node:child_process';
 import { fileURLToPath } from 'node:url';
+import { stageBundledIntegrationKeys, TMDB_API_KEY_ENV } from './bundled-integration-keys.mjs';
 
 const scriptDir = path.dirname(fileURLToPath(import.meta.url));
 const repoRoot = path.resolve(scriptDir, '..');
@@ -111,6 +112,17 @@ function buildMacosHelper() {
   }
 }
 
+// Only the key names are logged, never the values: CI masks secrets, but a
+// stray echo would still leak them into local build logs.
+function stageIntegrationKeys() {
+  const staged = stageBundledIntegrationKeys(path.join(repoRoot, 'dist'));
+  process.stdout.write(
+    staged.length > 0
+      ? `Staged bundled integration keys: ${staged.join(', ')}\n`
+      : `No bundled integration keys (${TMDB_API_KEY_ENV} unset)\n`,
+  );
+}
+
 function main() {
   fs.cpSync(path.join(rendererSourceDir, 'fonts'), path.join(repoRoot, 'dist', 'fonts'), {
     recursive: true,
@@ -121,6 +133,7 @@ function main() {
   copySyncUiAssets();
   copyAnimeUiAssets();
   buildMacosHelper();
+  stageIntegrationKeys();
 }
 
 main();

scripts/run-coverage-lane.test.ts

@@ -1,8 +1,10 @@
 import assert from 'node:assert/strict';
-import { resolve } from 'node:path';
+import { mkdirSync, mkdtempSync, rmSync, writeFileSync } from 'node:fs';
+import { tmpdir } from 'node:os';
+import { join, resolve } from 'node:path';
 import test from 'node:test';
 
-import { mergeLcovReports, resolveCoverageDir } from './run-coverage-lane';
+import { mergeLcovReports, resolveCoverageDir, runCoverageLane } from './run-coverage-lane';
 
 test('mergeLcovReports combines duplicate source-file counters across shard outputs', () => {
   const merged = mergeLcovReports([
@@ -72,3 +74,33 @@ test('resolveCoverageDir keeps coverage output inside the repository', () => {
   assert.throws(() => resolveCoverageDir(repoRoot, ['--coverage-dir', '../escape']));
   assert.throws(() => resolveCoverageDir(repoRoot, ['--coverage-dir', '/tmp/escape']));
 });
+
+test('runCoverageLane returns a failure when a discovered test fails', () => {
+  const repoRoot = mkdtempSync(join(tmpdir(), 'subminer-coverage-failure-'));
+  try {
+    mkdirSync(join(repoRoot, 'src'));
+    writeFileSync(
+      join(repoRoot, 'src', 'failure.test.ts'),
+      [
+        "import assert from 'node:assert/strict';",
+        "import test from 'node:test';",
+        '',
+        "test('intentional coverage failure', () => {",
+        "  assert.fail('coverage runner must propagate this failure');",
+        '});',
+        '',
+      ].join('\n'),
+    );
+
+    assert.notEqual(
+      runCoverageLane({
+        repoRootDir: repoRoot,
+        argv: ['bun-src-full', '--coverage-dir', 'coverage/test-src'],
+        stdio: 'pipe',
+      }),
+      0,
+    );
+  } finally {
+    rmSync(repoRoot, { recursive: true, force: true });
+  }
+});

scripts/run-coverage-lane.ts

@@ -201,14 +201,18 @@ export function mergeLcovReports(reports: string[]): string {
   return chunks.length > 0 ? `${chunks.join('\n')}\n` : '';
 }
 
-function runCoverageLane(): number {
-  const laneName = process.argv[2];
+export function runCoverageLane(
+  options: { repoRootDir?: string; argv?: string[]; stdio?: 'inherit' | 'pipe' } = {},
+): number {
+  const repoRootDir = options.repoRootDir ?? repoRoot;
+  const argv = options.argv ?? process.argv.slice(2);
+  const laneName = argv[0];
   if (laneName === undefined) {
     process.stderr.write('Missing coverage lane name\n');
     return 1;
   }
 
-  const coverageDir = resolveCoverageDir(repoRoot, process.argv.slice(3));
+  const coverageDir = resolveCoverageDir(repoRootDir, argv.slice(1));
   const shardRoot = join(coverageDir, '.shards');
   mkdirSync(coverageDir, { recursive: true });
   rmSync(shardRoot, { recursive: true, force: true });
@@ -216,7 +220,7 @@ function runCoverageLane(): number {
 
   let files: string[];
   try {
-    files = collectLaneFiles(repoRoot, laneName);
+    files = collectLaneFiles(repoRootDir, laneName);
   } catch (error) {
     process.stderr.write(`${error instanceof Error ? error.message : error}\n`);
     return 1;
@@ -230,8 +234,8 @@ function runCoverageLane(): number {
         'bun',
         ['test', '--coverage', '--coverage-reporter=lcov', '--coverage-dir', shardDir, `./${file}`],
         {
-          cwd: repoRoot,
-          stdio: 'inherit',
+          cwd: repoRootDir,
+          stdio: options.stdio ?? 'inherit',
         },
       );
 
@@ -253,7 +257,7 @@ function runCoverageLane(): number {
 
     writeFileSync(join(coverageDir, 'lcov.info'), mergeLcovReports(reports), 'utf8');
     process.stdout.write(
-      `Merged LCOV written to ${relative(repoRoot, join(coverageDir, 'lcov.info'))}\n`,
+      `Merged LCOV written to ${relative(repoRootDir, join(coverageDir, 'lcov.info'))}\n`,
     );
     return 0;
   } finally {

src/anime-bridge/playback-outcome.test.ts

@@ -20,6 +20,8 @@ function createHarness(overrides?: {
   timeoutMs?: number;
   probeIntervalMs?: number;
   readCostMs?: number;
+  onWait?: (elapsedMs: number, harness: Harness) => void;
+  isCurrent?: () => boolean;
 }) {
   const listeners = new Set<(event: PlaybackEndFileEvent) => void>();
   const properties = new Map<string, unknown>();
@@ -39,10 +41,12 @@ function createHarness(overrides?: {
     },
     wait: async (ms) => {
       clock += ms;
+      overrides?.onWait?.(clock, harness);
     },
     now: () => clock,
     timeoutMs: overrides?.timeoutMs ?? 1000,
     probeIntervalMs: overrides?.probeIntervalMs ?? 100,
+    isCurrent: overrides?.isCurrent,
   });
 
   const harness: Harness = {
@@ -94,6 +98,76 @@ test('times out with a failure when nothing ever starts', async () => {
   watch.dispose();
 });
 
+test('waits for a slow stream that is still loading after the confirmation deadline', async () => {
+  const { watch, harness } = createHarness({
+    timeoutMs: 300,
+    onWait: (elapsedMs, state) => {
+      if (elapsedMs >= 600) state.setProperty('vo-configured', true);
+    },
+  });
+  harness.setProperty('idle-active', false);
+  assert.deepEqual(await watch.wait(), { ok: true });
+  assert.equal(harness.elapsed(), 600);
+  watch.dispose();
+});
+
+test('reports a real stream error after the confirmation deadline', async () => {
+  const { watch, harness } = createHarness({
+    timeoutMs: 300,
+    onWait: (elapsedMs, state) => {
+      if (elapsedMs >= 600) state.emitEndFile({ reason: 'error', fileError: 'HTTP 503' });
+    },
+  });
+  harness.setProperty('idle-active', false);
+  assert.deepEqual(await watch.wait(), {
+    ok: false,
+    error: 'mpv could not play this stream: HTTP 503',
+  });
+  watch.dispose();
+});
+
+test('stops waiting if a slow stream leaves mpv idle', async () => {
+  const { watch, harness } = createHarness({
+    timeoutMs: 300,
+    onWait: (elapsedMs, state) => {
+      if (elapsedMs >= 600) state.setProperty('idle-active', true);
+    },
+  });
+  harness.setProperty('idle-active', false);
+  assert.equal((await watch.wait()).ok, false);
+  assert.equal(harness.elapsed(), 600);
+  watch.dispose();
+});
+
+test('stops watching a slow stream when the request is superseded', async () => {
+  let current = true;
+  const { watch, harness } = createHarness({
+    timeoutMs: 300,
+    isCurrent: () => current,
+    onWait: (elapsedMs) => {
+      if (elapsedMs >= 600) current = false;
+    },
+  });
+  harness.setProperty('idle-active', false);
+  assert.equal((await watch.wait()).ok, false);
+  assert.equal(harness.elapsed(), 600);
+  watch.dispose();
+  assert.equal(harness.listenerCount(), 0);
+});
+
+test('disposing the watcher stops polling a slow stream', async () => {
+  const { watch, harness } = createHarness({
+    timeoutMs: 300,
+    onWait: (elapsedMs) => {
+      if (elapsedMs >= 600) watch.dispose();
+    },
+  });
+  harness.setProperty('idle-active', false);
+  assert.equal((await watch.wait()).ok, false);
+  assert.equal(harness.elapsed(), 600);
+  assert.equal(harness.listenerCount(), 0);
+});
+
 test('slow property reads eat the budget instead of extending it', async () => {
   const { watch, harness } = createHarness({
     timeoutMs: 300,
@@ -102,7 +176,7 @@ test('slow property reads eat the budget instead of extending it', async () => {
   });
   const outcome = await watch.wait();
   assert.equal(outcome.ok, false);
-  // Two probes: 250 + 50 (the sleep clamped to what was left) then 250 again.
+  // Property reads and the final idle check all count toward elapsed time.
   assert.ok(harness.elapsed() >= 300, 'gave up before the timeout');
   assert.ok(harness.elapsed() < 900, 'read delays stretched the timeout');
   watch.dispose();

src/anime-bridge/playback-outcome.ts

@@ -26,10 +26,12 @@ export interface WatchPlaybackOutcomeDeps {
   /** One-shot mpv property read; may reject while the file is still loading. */
   readProperty: (name: string) => Promise<unknown>;
   wait: (ms: number) => Promise<void>;
-  /** Injectable clock; the timeout is wall-clock, not a probe count. */
+  /** Injectable clock for the initial confirmation deadline. */
   now?: () => number;
   timeoutMs?: number;
   probeIntervalMs?: number;
+  /** Stop watching when a newer episode replaces this request or the app closes. */
+  isCurrent?: () => boolean;
 }
 
 export interface PlaybackOutcomeWatch {
@@ -49,6 +51,7 @@ export function watchPlaybackOutcome(deps: WatchPlaybackOutcomeDeps): PlaybackOu
   const probeIntervalMs = deps.probeIntervalMs ?? DEFAULT_PROBE_INTERVAL_MS;
 
   let failure: PlaybackOutcome | null = null;
+  let disposed = false;
   const unsubscribe = deps.onEndFile((event) => {
     if (event.reason !== 'error') return;
     failure = {
@@ -60,11 +63,11 @@ export function watchPlaybackOutcome(deps: WatchPlaybackOutcomeDeps): PlaybackOu
   });
 
   async function wait(): Promise<PlaybackOutcome> {
-    // Wall-clock, not a probe count: a slow `readProperty` must eat into the
-    // budget rather than stretch it, and a zero probe interval must still end.
+    // Use elapsed time rather than probe count before checking whether mpv is
+    // still active. Slow property reads count toward this initial deadline.
     const now = deps.now ?? Date.now;
     const deadline = now() + timeoutMs;
-    while (now() < deadline) {
+    while (!disposed && (deps.isCurrent?.() ?? true)) {
       if (failure) return failure;
       try {
         if ((await deps.readProperty('vo-configured')) === true) return { ok: true };
@@ -72,10 +75,17 @@ export function watchPlaybackOutcome(deps: WatchPlaybackOutcomeDeps): PlaybackOu
         // The property is unreadable while mpv is between files; keep polling.
       }
       if (failure) return failure;
-      // Sleeping past the deadline would only delay the timeout report.
+      // A deadline without video is not a failure while mpv is still opening
+      // the stream. Keep waiting for video or a real end-file error in that case.
       const remaining = deadline - now();
-      if (remaining <= 0) break;
-      await deps.wait(Math.min(probeIntervalMs, remaining));
+      if (remaining <= 0) {
+        try {
+          if ((await deps.readProperty('idle-active')) !== false) break;
+        } catch {
+          break;
+        }
+      }
+      await deps.wait(remaining > 0 ? Math.min(probeIntervalMs, remaining) : probeIntervalMs);
     }
     return (
       failure ?? {
@@ -85,5 +95,11 @@ export function watchPlaybackOutcome(deps: WatchPlaybackOutcomeDeps): PlaybackOu
     );
   }
 
-  return { wait, dispose: unsubscribe };
+  return {
+    wait,
+    dispose: () => {
+      disposed = true;
+      unsubscribe();
+    },
+  };
 }

src/anki-integration.test.ts

@@ -1527,3 +1527,21 @@ test('AnkiIntegration.formatMiscInfoPattern avoids leaking Jellyfin api_key quer
   assert.equal(result, '[SubMiner] [Jellyfin/direct] Bocchi the Rock! - S01E02 (00:07:06)');
   assert.equal(result.includes('api_key='), false);
 });
+
+test('Anki metadata rejects a credential-bearing media title before metadata arrives', () => {
+  const integration = new AnkiIntegration(
+    { metadata: { pattern: '[SubMiner] %f | %F (%t)' } } as never,
+    {} as never,
+    {
+      currentVideoPath: 'https://jellyfin.example/Videos/item/stream?api_key=test-secret',
+      currentMediaTitle: 'stream?static=true&api_key=test-secret',
+      currentTimePos: 426,
+      send: () => true,
+    } as never,
+  );
+  const privateApi = integration as unknown as {
+    formatMiscInfoPattern: (fallbackFilename: string, startTimeSeconds?: number) => string;
+  };
+  const result = privateApi.formatMiscInfoPattern('stream?api_key=test-secret', 426);
+  assert.equal(result, '[SubMiner] Unknown media | Unknown media (00:07:06)');
+});

src/anki-integration.ts

@@ -32,6 +32,7 @@ import {
   type MediaTimingReviewRequest,
 } from './types/anki';
 import { AiConfig } from './types/integrations';
+import { sanitizeMediaTitle } from './shared/media-identity';
 import type { KnownWordMaturityTier } from './types/subtitle';
 import { MpvClient } from './types/runtime';
 import { OPEN_ANKI_CARD_ACTION_ID } from './types/notification';
@@ -1117,9 +1118,11 @@ export class AnkiIntegration {
       return null;
     }
     const mediaRange = this.getSubtitleMediaRange(context);
-    const timestamp = context
-      ? mediaRange.startTime + (mediaRange.endTime - mediaRange.startTime) / 2
-      : this.mpvClient.currentTimePos || 0;
+    const timestamp =
+      context?.screenshotTime ??
+      (context
+        ? mediaRange.startTime + (mediaRange.endTime - mediaRange.startTime) / 2
+        : this.mpvClient.currentTimePos || 0);
 
     if (this.config.media?.imageType === 'avif') {
       return this.mediaGenerator.generateAnimatedImage(
@@ -1165,11 +1168,13 @@ export class AnkiIntegration {
     }
 
     const videoFilename = extractFilenameFromMediaPath(mediaPath);
-    const resolvedMediaTitle = trimToNonEmptyString(mediaTitle);
+    const resolvedMediaTitle = sanitizeMediaTitle(mediaTitle);
     const filenameWithExt =
       (shouldPreferMediaTitleForMiscInfo(mediaPath, videoFilename)
-        ? resolvedMediaTitle || videoFilename
-        : videoFilename || resolvedMediaTitle) || fallbackFilename;
+        ? resolvedMediaTitle || 'Unknown media'
+        : sanitizeMediaTitle(videoFilename) || resolvedMediaTitle) ||
+      sanitizeMediaTitle(fallbackFilename) ||
+      'Unknown media';
     const filenameWithoutExt = filenameWithExt.replace(/\.[^.]+$/, '');
 
     const currentTimePos =
@@ -1794,6 +1799,8 @@ export class AnkiIntegration {
       ...request,
       audioPadding: Math.max(0, this.config.media.audioPadding ?? 0),
       maxMediaDuration: Math.max(0, this.config.media.maxMediaDuration ?? 30),
+      screenshotEnabled:
+        this.config.media.generateImage !== false && this.config.media.imageType !== 'avif',
     });
   }
 

src/anki-integration/animated-image-sync.test.ts

@@ -14,7 +14,8 @@ test('resolveAnimatedImageLeadInSeconds sums configured word audio durations for
   const leadInSeconds = await resolveAnimatedImageLeadInSeconds({
     config: {
       fields: {
-        audio: 'ExpressionAudio',
+        audio: 'SentenceAudio',
+        wordAudio: 'Pronunciation',
       },
       media: {
         imageType: 'avif',
@@ -25,7 +26,8 @@ test('resolveAnimatedImageLeadInSeconds sums configured word audio durations for
     noteInfo: {
       noteId: 42,
       fields: {
-        ExpressionAudio: {
+        SentenceAudio: { value: '[sound:sentence.mp3]' },
+        Pronunciation: {
           value: '[sound:word.mp3][sound:alt.ogg]',
         },
       },
@@ -121,3 +123,32 @@ test('resolveAnimatedImageLeadInSeconds falls back to zero when sync is disabled
 
   assert.equal(leadInSeconds, 0);
 });
+
+for (const sentenceAudio of ['', '[sound:sentence.mp3]']) {
+  test(`word audio defaults independently of sentence audio (${sentenceAudio ? 'existing' : 'new'} note)`, async () => {
+    const retrieved: string[] = [];
+    const leadInSeconds = await resolveAnimatedImageLeadInSeconds({
+      config: {
+        fields: { audio: 'SentenceAudio' },
+        media: { imageType: 'avif' },
+      },
+      noteInfo: {
+        noteId: 42,
+        fields: {
+          ExpressionAudio: { value: '[sound:word.mp3]' },
+          SentenceAudio: { value: sentenceAudio },
+        },
+      },
+      resolveConfiguredFieldName: (noteInfo, ...preferredNames) =>
+        preferredNames.find((name) => name !== undefined && name in noteInfo.fields) ?? null,
+      retrieveMediaFileBase64: async (filename) => {
+        retrieved.push(filename);
+        return 'd29yZA==';
+      },
+      probeAudioDurationSeconds: async (_buffer, filename) => (filename === 'word.mp3' ? 0.6 : 4),
+    });
+
+    assert.equal(leadInSeconds, 0.6);
+    assert.deepEqual(retrieved, ['word.mp3']);
+  });
+}

src/anki-integration/animated-image-sync.ts

@@ -97,8 +97,8 @@ export async function resolveAnimatedImageLeadInSeconds<TNoteInfo extends NoteIn
 
   const wordAudioFieldName = resolveConfiguredFieldName(
     noteInfo,
-    config.fields?.audio,
-    DEFAULT_ANKI_CONNECT_CONFIG.fields.audio,
+    config.fields?.wordAudio,
+    DEFAULT_ANKI_CONNECT_CONFIG.fields.wordAudio,
   );
   if (!wordAudioFieldName) {
     return 0;

src/anki-integration/card-creation-manual-update.test.ts

@@ -160,6 +160,52 @@ test('manual clipboard subtitle update replaces audio in the configured field',
   );
 });
 
+test('manual clipboard mining treats a zero media duration cap as unlimited', async () => {
+  const audioRanges: Array<{ start: number; end: number; padding: number | undefined }> = [];
+  const scenarios = [
+    { maxMediaDuration: 0, expectedEnd: 14 },
+    { maxMediaDuration: 1, expectedEnd: 13 },
+  ];
+
+  for (const scenario of scenarios) {
+    const { service } = createManualUpdateService({
+      getConfig: () =>
+        ({
+          deck: 'Mining',
+          fields: {
+            word: 'Expression',
+            sentence: 'Sentence',
+            audio: 'ExpressionAudio',
+          },
+          media: {
+            generateAudio: true,
+            generateImage: false,
+            audioPadding: 0.25,
+            maxMediaDuration: scenario.maxMediaDuration,
+          },
+          behavior: {},
+          ai: false,
+        }) as AnkiConnectConfig,
+      mediaGenerator: {
+        generateAudio: async (_path, start, end, padding) => {
+          audioRanges.push({ start, end, padding });
+          return Buffer.from('audio');
+        },
+        generateScreenshot: async () => null,
+        generateAnimatedImage: async () => null,
+      },
+    });
+
+    await service.updateLastAddedFromClipboard('字幕');
+
+    assert.deepEqual(audioRanges.at(-1), {
+      start: 12,
+      end: scenario.expectedEnd,
+      padding: 0.25,
+    });
+  }
+});
+
 test('manual clipboard word-card update uses configured fields with Lapis and Kiku enabled', async () => {
   const { service, updatedFields } = createManualUpdateService({
     getConfig: () =>

src/anki-integration/card-creation.test.ts

@@ -290,6 +290,9 @@ test('CardCreationService keeps updating after recordCardsMinedCallback throws',
 });
 
 test('CardCreationService uses stream-open-filename for remote media generation', async () => {
+  let reviewing = false;
+  const audioRanges: number[][] = [];
+  const imageTimes: number[] = [];
   const audioPaths: string[] = [];
   const imagePaths: string[] = [];
   const recordMediaPath = (mediaInput: MediaInput): string =>
@@ -319,6 +322,10 @@ test('CardCreationService uses stream-open-filename for remote media generation'
         behavior: {},
         ai: false,
       }) as AnkiConnectConfig,
+    reviewMediaTiming: async () =>
+      reviewing
+        ? { action: 'confirm', startTime: 0.2, endTime: 0.8, screenshotTime: 3.125 }
+        : { action: 'use-original' },
     getAiConfig: () => ({}),
     getTimingTracker: () => ({}) as never,
     getMpvClient: () =>
@@ -349,16 +356,18 @@ test('CardCreationService uses stream-open-filename for remote media generation'
       ],
       updateNoteFields: async () => undefined,
       storeMediaFile: async () => undefined,
-      findNotes: async () => [],
+      findNotes: async () => [42],
       retrieveMediaFile: async () => '',
       deleteNotes: async () => undefined,
     },
     mediaGenerator: {
-      generateAudio: async (path) => {
+      generateAudio: async (path, start, end, padding) => {
+        audioRanges.push([start, end, padding ?? -1]);
         audioPaths.push(recordMediaPath(path));
         return Buffer.from('audio');
       },
-      generateScreenshot: async (path) => {
+      generateScreenshot: async (path, timestamp) => {
+        imageTimes.push(timestamp);
         imagePaths.push(recordMediaPath(path));
         return Buffer.from('image');
       },
@@ -406,6 +415,14 @@ test('CardCreationService uses stream-open-filename for remote media generation'
   assert.equal(created, true);
   assert.deepEqual(audioPaths, [audioUrl]);
   assert.deepEqual(imagePaths, [videoUrl]);
+  reviewing = true;
+  assert.equal(await service.createSentenceCard('テスト', 0, 1), true);
+  assert.deepEqual(audioRanges.at(-1), [0.2, 0.8, 0]);
+  assert.equal(imageTimes.at(-1), 3.125);
+  await service.markLastCardAsAudioCard();
+  assert.equal(imageTimes.length, 3);
+  assert.deepEqual(audioRanges.at(-1), [0.2, 0.8, 0]);
+  assert.equal(imageTimes.at(-1), 3.125);
 });
 
 test('CardCreationService does not use mpv stream indexes for ready cached YouTube media', async () => {

src/anki-integration/card-creation.ts

@@ -21,6 +21,7 @@ import {
   resolveAudioStreamIndexForMediaGeneration,
   type MediaGenerationInputResolverOptions,
 } from './media-source';
+import { clampMediaEndTime } from './media-duration';
 import { resolveWordCardKind } from './note-field-utils';
 import type { PendingYoutubeMediaUpdate } from './pending-youtube-media';
 import { resolveMpvVolumeScale } from './mpv-volume';
@@ -233,11 +234,12 @@ export class CardCreationService {
       let rangeEnd = Math.max(...timings.map((entry) => entry.endTime));
 
       const maxMediaDuration = this.deps.getConfig().media?.maxMediaDuration ?? 30;
-      if (maxMediaDuration > 0 && rangeEnd - rangeStart > maxMediaDuration) {
+      const cappedRangeEnd = clampMediaEndTime(rangeStart, rangeEnd, maxMediaDuration);
+      if (cappedRangeEnd !== rangeEnd) {
         log.warn(
           `Media range ${(rangeEnd - rangeStart).toFixed(1)}s exceeds cap of ${maxMediaDuration}s, clamping`,
         );
-        rangeEnd = rangeStart + maxMediaDuration;
+        rangeEnd = cappedRangeEnd;
       }
 
       this.deps.showOsdNotification('Updating card from clipboard...');
@@ -437,9 +439,7 @@ export class CardCreationService {
       }
 
       const maxMediaDuration = this.deps.getConfig().media?.maxMediaDuration ?? 30;
-      if (maxMediaDuration > 0 && endTime - startTime > maxMediaDuration) {
-        endTime = startTime + maxMediaDuration;
-      }
+      endTime = clampMediaEndTime(startTime, endTime, maxMediaDuration);
 
       this.deps.showOsdNotification('Marking card as audio card...');
       await this.deps.withUpdateProgress('Marking audio card', async () => {
@@ -536,6 +536,7 @@ export class CardCreationService {
               endTime,
               animatedLeadInSeconds,
               exactReviewedRange,
+              timingDecision.action === 'confirm' ? timingDecision.screenshotTime : undefined,
             );
 
             const imageField = this.deps.getConfig().fields?.image;
@@ -599,11 +600,12 @@ export class CardCreationService {
     }
 
     const maxMediaDuration = this.deps.getConfig().media?.maxMediaDuration ?? 30;
-    if (maxMediaDuration > 0 && endTime - startTime > maxMediaDuration) {
+    const cappedEndTime = clampMediaEndTime(startTime, endTime, maxMediaDuration);
+    if (cappedEndTime !== endTime) {
       log.warn(
         `Sentence card media range ${(endTime - startTime).toFixed(1)}s exceeds cap of ${maxMediaDuration}s, clamping`,
       );
-      endTime = startTime + maxMediaDuration;
+      endTime = cappedEndTime;
     }
 
     try {
@@ -796,6 +798,9 @@ export class CardCreationService {
             generateImage,
             volumeScale,
             ...(exactReviewedRange ? { mediaPaddingSeconds: 0 } : {}),
+            ...(timingDecision.action === 'confirm' && timingDecision.screenshotTime !== undefined
+              ? { screenshotTime: timingDecision.screenshotTime }
+              : {}),
           });
           await this.deps.showNotification(noteId, label, 'media queued');
           return true;
@@ -840,6 +845,7 @@ export class CardCreationService {
               endTime,
               0,
               exactReviewedRange,
+              timingDecision.action === 'confirm' ? timingDecision.screenshotTime : undefined,
             );
 
             const imageField = config.fields?.image;
@@ -922,15 +928,16 @@ export class CardCreationService {
     endTime: number,
     animatedLeadInSeconds = 0,
     exactReviewedRange = false,
+    screenshotTime?: number,
   ): Promise<Buffer | null> {
     const mpvClient = this.deps.getMpvClient();
     if (!mpvClient) {
       return null;
     }
 
-    const timestamp = exactReviewedRange
-      ? startTime + (endTime - startTime) / 2
-      : mpvClient.currentTimePos || 0;
+    const timestamp =
+      screenshotTime ??
+      (exactReviewedRange ? startTime + (endTime - startTime) / 2 : mpvClient.currentTimePos || 0);
 
     if (this.deps.getConfig().media?.imageType === 'avif') {
       let imageStart = startTime;

src/anki-integration/media-duration.ts

@@ -0,0 +1,10 @@
+/** Zero or a negative cap leaves the requested end time unchanged. */
+export function clampMediaEndTime(
+  startTime: number,
+  endTime: number,
+  maxMediaDuration: number,
+): number {
+  return maxMediaDuration > 0 && endTime - startTime > maxMediaDuration
+    ? startTime + maxMediaDuration
+    : endTime;
+}

src/anki-integration/note-update-workflow.test.ts

@@ -729,6 +729,7 @@ test('NoteUpdateWorkflow uses the combined review sentence for the card and medi
     startTime: 2,
     endTime: 7,
     text: 'previous-line current-line next-line',
+    screenshotTime: 8,
   });
   harness.deps.generateAudio = async (context) => {
     audioContexts.push(context);
@@ -745,6 +746,7 @@ test('NoteUpdateWorkflow uses the combined review sentence for the card and medi
   assert.equal(audioContexts[0]?.startTime, 2);
   assert.equal(audioContexts[0]?.endTime, 7);
   assert.equal(audioContexts[0]?.mediaPaddingSeconds, 0);
+  assert.equal(audioContexts[0]?.screenshotTime, 8);
 });
 
 test('NoteUpdateWorkflow keeps cache unchanged and reports when deletion fails', async () => {

src/anki-integration/note-update-workflow.ts

@@ -253,6 +253,9 @@ export class NoteUpdateWorkflow {
             startTime: timingDecision.startTime,
             endTime: timingDecision.endTime,
             mediaPaddingSeconds: 0,
+            ...(timingDecision.screenshotTime !== undefined
+              ? { screenshotTime: timingDecision.screenshotTime }
+              : {}),
           };
         } else if (timingDecision.action === 'skip-media') {
           skipMedia = true;

src/anki-integration/pending-youtube-media-queue.test.ts

@@ -51,6 +51,41 @@ function createDeps(
   return deps;
 }
 
+test('queued media keeps a chosen screenshot separate from the reviewed audio range', async () => {
+  const screenshots: number[] = [];
+  const audioRanges: number[][] = [];
+  const deps = createDeps();
+  deps.client.notesInfo = async () => [{ noteId: 42, fields: { Picture: { value: '' } } }];
+  deps.mediaGenerator.generateScreenshot = async (_media, time) => {
+    screenshots.push(time);
+    return Buffer.from('image');
+  };
+  deps.mediaGenerator.generateAudio = async (_media, start, end, padding) => {
+    audioRanges.push([start, end, padding ?? -1]);
+    return Buffer.from('audio');
+  };
+  const queue = new PendingYoutubeMediaQueue(deps);
+  assert.equal(
+    await queue.queueFromNote({
+      noteId: 42,
+      noteInfo: { noteId: 42, fields: {} },
+      label: 'test',
+      context: {
+        source: 'overlay',
+        text: '字幕',
+        startTime: 1,
+        endTime: 2,
+        mediaPaddingSeconds: 0,
+        screenshotTime: 3.125,
+      },
+    }),
+    true,
+  );
+  await queue.handleReady('https://youtu.be/abc123', '/cache/video.mkv');
+  assert.deepEqual(screenshots, [3.125]);
+  assert.deepEqual(audioRanges, [[1, 2, 0]]);
+});
+
 test('PendingYoutubeMediaQueue treats cache lookup failures as an immediate generation fallback', async () => {
   const deps = createDeps({
     getCachedMediaPath: async () => {

src/anki-integration/pending-youtube-media-queue.ts

@@ -147,6 +147,9 @@ export class PendingYoutubeMediaQueue {
       generateAudio: shouldGenerateAudio(config),
       generateImage: shouldGenerateImage(config),
       volumeScale,
+      ...(job.context?.screenshotTime !== undefined
+        ? { screenshotTime: job.context.screenshotTime }
+        : {}),
       ...(job.context?.mediaPaddingSeconds !== undefined
         ? { mediaPaddingSeconds: job.context.mediaPaddingSeconds }
         : {}),
@@ -320,6 +323,7 @@ export class PendingYoutubeMediaQueue {
           job.endTime,
           animatedLeadInSeconds,
           job.mediaPaddingSeconds,
+          job.screenshotTime,
         );
         if (imageBuffer) {
           await this.deps.client.storeMediaFile(imageFilename, imageBuffer);
@@ -381,6 +385,7 @@ export class PendingYoutubeMediaQueue {
     endTime: number,
     animatedLeadInSeconds = 0,
     mediaPaddingSeconds?: number,
+    screenshotTime?: number,
   ): Promise<Buffer | null> {
     const config = this.deps.getConfig();
     if (config.media?.imageType === 'avif') {
@@ -399,7 +404,7 @@ export class PendingYoutubeMediaQueue {
       );
     }
 
-    const timestamp = startTime + (endTime - startTime) / 2;
+    const timestamp = screenshotTime ?? startTime + (endTime - startTime) / 2;
     return this.deps.mediaGenerator.generateScreenshot(videoPath, timestamp, {
       format: config.media?.imageFormat as 'jpg' | 'png' | 'webp',
       quality: config.media?.imageQuality,

src/anki-integration/pending-youtube-media.ts

@@ -11,6 +11,7 @@ export interface PendingYoutubeMediaUpdate {
   generateImage: boolean;
   volumeScale?: number;
   mediaPaddingSeconds?: number;
+  screenshotTime?: number;
 }
 
 function trimToNonEmptyString(value: unknown): string | null {

src/ci-workflow.test.ts

@@ -19,6 +19,23 @@ test('package scripts expose a sharded maintained source coverage lane with lcov
   );
 });
 
+test('source and coverage scripts discover the same maintained source lane', () => {
+  const sourceLane = packageJson.scripts['test:src']?.match(/run-test-lane\.mjs\s+([^\s]+)/)?.[1];
+  const coverageLane = packageJson.scripts['test:coverage:src']?.match(
+    /run-coverage-lane\.ts\s+([^\s]+)/,
+  )?.[1];
+
+  assert.equal(sourceLane, 'bun-src-full');
+  assert.equal(coverageLane, sourceLane);
+});
+
+test('environment suite owns launcher smoke execution', () => {
+  assert.match(
+    packageJson.scripts['test:env'] ?? '',
+    /^bun run test:launcher:smoke:src && bun run test:plugin:src && bun run test:immersion:sqlite:src$/,
+  );
+});
+
 test('ci delegates its gate instead of duplicating quality steps', () => {
   assert.match(
     ciWorkflow,

src/config/config.test.ts

@@ -2819,6 +2819,29 @@ test('forces Senren off when Kiku is also enabled and validates Senren fieldGrou
   );
 });
 
+test('warns and falls back when isSenren.enabled is not boolean', () => {
+  const dir = makeTempDir();
+  fs.writeFileSync(
+    path.join(dir, 'config.jsonc'),
+    `{
+      "ankiConnect": {
+        "isSenren": { "enabled": "true" }
+      }
+    }`,
+    'utf-8',
+  );
+
+  const service = new ConfigService(dir);
+
+  assert.equal(
+    service.getConfig().ankiConnect.isSenren.enabled,
+    DEFAULT_CONFIG.ankiConnect.isSenren.enabled,
+  );
+  assert.ok(
+    service.getWarnings().some((warning) => warning.path === 'ankiConnect.isSenren.enabled'),
+  );
+});
+
 test('accepts valid ankiConnect knownWords deck object', () => {
   const dir = makeTempDir();
   fs.writeFileSync(

src/config/definitions.ts

@@ -43,6 +43,7 @@ const {
   anime,
   jimaku,
   tsukihime,
+  tmdb,
   anilist,
   mpv,
   yomitan,
@@ -78,6 +79,7 @@ export const DEFAULT_CONFIG: ResolvedConfig = {
   anime,
   jimaku,
   tsukihime,
+  tmdb,
   anilist,
   mpv,
   yomitan,

src/config/definitions/defaults-integrations.ts

@@ -7,6 +7,7 @@ export const INTEGRATIONS_DEFAULT_CONFIG: Pick<
   | 'anime'
   | 'jimaku'
   | 'tsukihime'
+  | 'tmdb'
   | 'anilist'
   | 'mpv'
   | 'yomitan'
@@ -30,6 +31,7 @@ export const INTEGRATIONS_DEFAULT_CONFIG: Pick<
     fields: {
       word: 'Expression',
       audio: 'ExpressionAudio',
+      wordAudio: 'ExpressionAudio',
       image: 'Picture',
       sentence: 'Sentence',
       miscInfo: 'MiscInfo',
@@ -123,6 +125,10 @@ export const INTEGRATIONS_DEFAULT_CONFIG: Pick<
     apiBaseUrl: 'https://api.tsukihime.org/v1',
     maxSearchResults: 10,
   },
+  tmdb: {
+    apiKey: '',
+    apiKeyCommand: '',
+  },
   mpv: {
     executablePath: '',
     launchMode: 'normal',

src/config/definitions/options-integrations.ts

@@ -82,6 +82,13 @@ export function buildIntegrationConfigOptionRegistry(
       defaultValue: defaultConfig.ankiConnect.fields.audio,
       description: 'Card field that receives generated sentence audio.',
     },
+    {
+      path: 'ankiConnect.fields.wordAudio',
+      kind: 'string',
+      defaultValue: defaultConfig.ankiConnect.fields.wordAudio,
+      description:
+        'Existing word-audio field read to time the frozen first frame of animated images. This mapping is only used for synchronization.',
+    },
     {
       path: 'ankiConnect.fields.image',
       kind: 'string',
@@ -288,7 +295,7 @@ export function buildIntegrationConfigOptionRegistry(
       path: 'ankiConnect.media.maxMediaDuration',
       kind: 'number',
       defaultValue: defaultConfig.ankiConnect.media.maxMediaDuration,
-      description: 'Maximum allowed media clip duration in seconds.',
+      description: 'Maximum allowed media clip duration in seconds. 0 disables the cap.',
     },
     {
       path: 'ankiConnect.knownWords.matchMode',
@@ -472,6 +479,20 @@ export function buildIntegrationConfigOptionRegistry(
       defaultValue: defaultConfig.tsukihime.maxSearchResults,
       description: 'Maximum TsukiHime search results returned.',
     },
+    {
+      path: 'tmdb.apiKey',
+      kind: 'string',
+      defaultValue: defaultConfig.tmdb.apiKey,
+      description:
+        'Your own TMDB API key or read access token for live-action posters and synopses in the stats Library. Release builds bundle a project key, so set this only to use your own quota or when running from source (free under Settings > API on themoviedb.org).',
+    },
+    {
+      path: 'tmdb.apiKeyCommand',
+      kind: 'string',
+      defaultValue: defaultConfig.tmdb.apiKeyCommand,
+      description:
+        'Shell command that prints the TMDB API key to stdout. Used instead of apiKey to avoid storing the key in plain text.',
+    },
     {
       path: 'anilist.enabled',
       kind: 'boolean',

src/config/definitions/template-sections.ts

@@ -144,7 +144,7 @@ const INTEGRATION_TEMPLATE_SECTIONS: ConfigTemplateSection[] = [
     title: 'AnkiConnect Integration',
     description: ['Automatic Anki updates and media generation options.'],
     notes: [
-      'Hot-reload: ankiConnect.ai.enabled, media.normalizeAudio/mirrorMpvVolume/reviewTiming, knownWords, nPlusOne, fields.word/audio/image/sentence/miscInfo, behavior.autoUpdateNewCards, isLapis.sentenceCardModel, isKiku.fieldGrouping, isSenren.fieldGrouping, and lapisKiku.wordCardKind update live while SubMiner is running.',
+      'Hot-reload: ankiConnect.ai.enabled, media.normalizeAudio/mirrorMpvVolume/reviewTiming, knownWords, nPlusOne, fields.word/audio/wordAudio/image/sentence/miscInfo, behavior.autoUpdateNewCards, isLapis.sentenceCardModel, isKiku.fieldGrouping, isSenren.fieldGrouping, and lapisKiku.wordCardKind update live while SubMiner is running.',
       'Shared AI provider transport settings are read from top-level ai and typically require restart.',
       'Most other AnkiConnect settings still require restart.',
     ],
@@ -175,6 +175,14 @@ const INTEGRATION_TEMPLATE_SECTIONS: ConfigTemplateSection[] = [
     notes: ['Hot-reload: TsukiHime changes apply to the next TsukiHime request.'],
     key: 'tsukihime',
   },
+  {
+    title: 'TMDB',
+    description: [
+      'TMDB (The Movie Database) metadata for live-action dramas and movies in the stats Library: posters, synopses, and grouping by show.',
+    ],
+    notes: ['Hot-reload: TMDB changes apply to the next TMDB request.'],
+    key: 'tmdb',
+  },
   {
     title: 'YouTube Playback Settings',
     description: [

src/config/hot-reload.ts

@@ -0,0 +1,66 @@
+function pathStartsWith(path: string, prefix: string): boolean {
+  return path === prefix || path.startsWith(`${prefix}.`);
+}
+
+const HOT_RELOAD_ROOTS = ['subtitleStyle', 'keybindings', 'shortcuts', 'subtitleSidebar'] as const;
+
+const HOT_RELOAD_EXACT_OR_PREFIX_PATHS = [
+  'secondarySub.defaultMode',
+  'mpv.aniskipEnabled',
+  'mpv.aniskipButtonKey',
+  'ankiConnect.ai.enabled',
+  'stats.toggleKey',
+  'stats.markWatchedKey',
+  'logging.level',
+  'logging.rotation',
+  'logging.files',
+  'youtube.primarySubLanguages',
+  'anime.autoOpenJimaku',
+  'ankiConnect.deck',
+  'ankiConnect.media.normalizeAudio',
+  'ankiConnect.media.mirrorMpvVolume',
+  'ankiConnect.media.reviewTiming',
+  'ankiConnect.behavior.autoUpdateNewCards',
+  'ankiConnect.knownWords.highlightEnabled',
+  'ankiConnect.knownWords.refreshMinutes',
+  'ankiConnect.knownWords.addMinedWordsImmediately',
+  'ankiConnect.knownWords.matchMode',
+  'ankiConnect.knownWords.decks',
+  'ankiConnect.nPlusOne.enabled',
+  'ankiConnect.nPlusOne.minSentenceWords',
+  'ankiConnect.fields.word',
+  'ankiConnect.fields.audio',
+  'ankiConnect.fields.wordAudio',
+  'ankiConnect.fields.image',
+  'ankiConnect.fields.sentence',
+  'ankiConnect.fields.miscInfo',
+  'ankiConnect.isLapis.sentenceCardModel',
+  'ankiConnect.isKiku.fieldGrouping',
+  'ankiConnect.isSenren.fieldGrouping',
+  'ankiConnect.lapisKiku.wordCardKind',
+] as const;
+
+export function getConfigHotReloadField(path: string): string | null {
+  for (const root of HOT_RELOAD_ROOTS) {
+    if (pathStartsWith(path, root)) {
+      return root;
+    }
+  }
+
+  for (const hotPath of HOT_RELOAD_EXACT_OR_PREFIX_PATHS) {
+    if (pathStartsWith(path, hotPath)) {
+      return hotPath;
+    }
+  }
+
+  // These consumers read the current config when the next operation starts.
+  if (
+    ['jimaku', 'tmdb', 'subsync', 'notifications', 'subtitleGeneration'].some((root) =>
+      pathStartsWith(path, root),
+    )
+  ) {
+    return path;
+  }
+
+  return null;
+}

src/config/resolve/anki-connect.test.ts

@@ -32,6 +32,101 @@ test('media timing review is disabled by default and accepts a boolean override'
   assert.deepEqual(enabledContext.warnings, []);
 });
 
+test('invalid direct and field-grouping Anki values warn and keep defaults', () => {
+  const { context, warnings } = makeContext({
+    enabled: 'true',
+    url: 8765,
+    pollingRate: '3000',
+    deck: ['Mining'],
+    isKiku: {
+      enabled: 1,
+      fieldGrouping: 'sometimes',
+      deleteDuplicateInAuto: 'false',
+    },
+    isSenren: {
+      enabled: 'true',
+      fieldGrouping: false,
+      deleteDuplicateInAuto: 0,
+    },
+  });
+
+  applyAnkiConnectResolution(context);
+
+  assert.equal(context.resolved.ankiConnect.enabled, DEFAULT_CONFIG.ankiConnect.enabled);
+  assert.equal(context.resolved.ankiConnect.url, DEFAULT_CONFIG.ankiConnect.url);
+  assert.equal(context.resolved.ankiConnect.pollingRate, DEFAULT_CONFIG.ankiConnect.pollingRate);
+  assert.equal(context.resolved.ankiConnect.deck, DEFAULT_CONFIG.ankiConnect.deck);
+  assert.deepEqual(context.resolved.ankiConnect.isKiku, DEFAULT_CONFIG.ankiConnect.isKiku);
+  assert.deepEqual(context.resolved.ankiConnect.isSenren, DEFAULT_CONFIG.ankiConnect.isSenren);
+  assert.deepEqual(
+    warnings.map((warning) => warning.path),
+    [
+      'ankiConnect.enabled',
+      'ankiConnect.url',
+      'ankiConnect.pollingRate',
+      'ankiConnect.deck',
+      'ankiConnect.isKiku.enabled',
+      'ankiConnect.isKiku.deleteDuplicateInAuto',
+      'ankiConnect.isKiku.fieldGrouping',
+      'ankiConnect.isSenren.enabled',
+      'ankiConnect.isSenren.deleteDuplicateInAuto',
+      'ankiConnect.isSenren.fieldGrouping',
+    ],
+  );
+});
+
+test('accepts valid direct and field-grouping Anki values', () => {
+  const { context, warnings } = makeContext({
+    enabled: false,
+    url: 'http://127.0.0.1:9876',
+    pollingRate: 750,
+    deck: 'Mining',
+    isKiku: {
+      enabled: true,
+      fieldGrouping: 'manual',
+      deleteDuplicateInAuto: false,
+    },
+    isSenren: {
+      enabled: false,
+      fieldGrouping: 'disabled',
+      deleteDuplicateInAuto: false,
+    },
+  });
+
+  applyAnkiConnectResolution(context);
+
+  assert.equal(context.resolved.ankiConnect.enabled, false);
+  assert.equal(context.resolved.ankiConnect.url, 'http://127.0.0.1:9876');
+  assert.equal(context.resolved.ankiConnect.pollingRate, 750);
+  assert.equal(context.resolved.ankiConnect.deck, 'Mining');
+  assert.deepEqual(context.resolved.ankiConnect.isKiku, {
+    enabled: true,
+    fieldGrouping: 'manual',
+    deleteDuplicateInAuto: false,
+  });
+  assert.deepEqual(context.resolved.ankiConnect.isSenren, {
+    enabled: false,
+    fieldGrouping: 'disabled',
+    deleteDuplicateInAuto: false,
+  });
+  assert.deepEqual(warnings, []);
+});
+
+test('ignores unknown Anki keys without warning or admitting them to resolved config', () => {
+  const { context, warnings } = makeContext({
+    futureOption: { enabled: true },
+    isKiku: { futureGroupingOption: 'future' },
+    isSenren: { futureGroupingOption: 'future' },
+  });
+
+  applyAnkiConnectResolution(context);
+
+  assert.equal(Object.hasOwn(context.resolved.ankiConnect, 'futureOption'), false);
+  assert.equal(Object.hasOwn(context.resolved.ankiConnect.isKiku, 'futureGroupingOption'), false);
+  assert.equal(Object.hasOwn(context.resolved.ankiConnect.isSenren, 'futureGroupingOption'), false);
+  assert.deepEqual(warnings, []);
+});
+
 test('modern media duration accepts zero as the disabled cap sentinel', () => {
   const disabledCap = makeContext({ media: { maxMediaDuration: 0 } });
   applyAnkiConnectResolution(disabledCap.context);
@@ -290,6 +385,28 @@ test('accepts ankiConnect.media.syncAnimatedImageToWordAudio override', () => {
   );
 });
 
+test('word audio mapping defaults and validates independently of sentence audio', () => {
+  for (const wordAudio of [undefined, 'Pronunciation', 7]) {
+    const { context, warnings } = makeContext({
+      fields: {
+        audio: 'SentenceAudio',
+        ...(wordAudio !== undefined ? { wordAudio } : {}),
+      },
+    });
+    applyAnkiConnectResolution(context);
+
+    assert.equal(context.resolved.ankiConnect.fields.audio, 'SentenceAudio');
+    assert.equal(
+      context.resolved.ankiConnect.fields.wordAudio,
+      typeof wordAudio === 'string' ? wordAudio : DEFAULT_CONFIG.ankiConnect.fields.wordAudio,
+    );
+    assert.deepEqual(
+      warnings.map((warning) => warning.path),
+      typeof wordAudio === 'number' ? ['ankiConnect.fields.wordAudio'] : [],
+    );
+  }
+});
+
 test('invalid modern Anki subtrees warn and keep resolved defaults', () => {
   const { context, warnings } = makeContext({
     fields: { word: 7 },

src/config/resolve/anki-connect.ts

@@ -19,11 +19,11 @@ export function applyAnkiConnectResolution(context: ResolveContext): void {
   const media = isObject(ankiConnect.media) ? ankiConnect.media : {};
   const metadata = isObject(ankiConnect.metadata) ? ankiConnect.metadata : {};
 
-  initializeAnkiConnectResolution(context, ankiConnect);
+  initializeAnkiConnectResolution(context);
   applyAnkiModernResolution(context, ankiConnect, behavior, media);
   applyAnkiLegacyResolution(context, ankiConnect, behavior, fields, media, metadata);
   applyAnkiKnownWordsResolution(context, ankiConnect, behavior);
-  applyAnkiKikuResolution(context);
-  applyAnkiSenrenResolution(context);
+  applyAnkiKikuResolution(context, ankiConnect);
+  applyAnkiSenrenResolution(context, ankiConnect);
   applyAnkiLapisKikuResolution(context, ankiConnect);
 }

src/config/resolve/anki-connect/base.ts

@@ -0,0 +1,58 @@
+import { DEFAULT_CONFIG } from '../../definitions';
+import type { ResolveContext } from '../context';
+import { asBoolean, asString } from '../shared';
+import { applyModernValue, asPositiveNumber } from './modern-value';
+
+export function applyAnkiBaseResolution(
+  context: ResolveContext,
+  ankiConnect: Record<string, unknown>,
+): void {
+  applyModernValue(
+    context,
+    ankiConnect,
+    'enabled',
+    'ankiConnect.enabled',
+    asBoolean,
+    DEFAULT_CONFIG.ankiConnect.enabled,
+    (value) => {
+      context.resolved.ankiConnect.enabled = value;
+    },
+    'Expected boolean.',
+  );
+  applyModernValue(
+    context,
+    ankiConnect,
+    'url',
+    'ankiConnect.url',
+    asString,
+    DEFAULT_CONFIG.ankiConnect.url,
+    (value) => {
+      context.resolved.ankiConnect.url = value;
+    },
+    'Expected string.',
+  );
+  applyModernValue(
+    context,
+    ankiConnect,
+    'pollingRate',
+    'ankiConnect.pollingRate',
+    asPositiveNumber,
+    DEFAULT_CONFIG.ankiConnect.pollingRate,
+    (value) => {
+      context.resolved.ankiConnect.pollingRate = value;
+    },
+    'Expected positive number.',
+  );
+  applyModernValue(
+    context,
+    ankiConnect,
+    'deck',
+    'ankiConnect.deck',
+    asString,
+    DEFAULT_CONFIG.ankiConnect.deck,
+    (value) => {
+      context.resolved.ankiConnect.deck = value;
+    },
+    'Expected string.',
+  );
+}

src/config/resolve/anki-connect/field-grouping-config.ts

@@ -0,0 +1,53 @@
+import { DEFAULT_CONFIG } from '../../definitions';
+import type { ResolveContext } from '../context';
+import { asBoolean, isObject } from '../shared';
+import { applyModernValue } from './modern-value';
+
+type FieldGroupingConfigKey = 'isKiku' | 'isSenren';
+
+export function applyFieldGroupingConfigResolution(
+  context: ResolveContext,
+  ankiConnect: Record<string, unknown>,
+  key: FieldGroupingConfigKey,
+): void {
+  const source = ankiConnect[key];
+  if (!isObject(source)) {
+    if (source !== undefined) {
+      context.warn(
+        `ankiConnect.${key}`,
+        source,
+        DEFAULT_CONFIG.ankiConnect[key],
+        'Expected object.',
+      );
+    }
+    return;
+  }
+
+  for (const booleanKey of ['enabled', 'deleteDuplicateInAuto'] as const) {
+    applyModernValue(
+      context,
+      source,
+      booleanKey,
+      `ankiConnect.${key}.${booleanKey}`,
+      asBoolean,
+      DEFAULT_CONFIG.ankiConnect[key][booleanKey],
+      (value) => {
+        context.resolved.ankiConnect[key][booleanKey] = value;
+      },
+      'Expected boolean.',
+    );
+  }
+
+  applyModernValue(
+    context,
+    source,
+    'fieldGrouping',
+    `ankiConnect.${key}.fieldGrouping`,
+    (value) => (value === 'auto' || value === 'manual' || value === 'disabled' ? value : undefined),
+    DEFAULT_CONFIG.ankiConnect[key].fieldGrouping,
+    (value) => {
+      context.resolved.ankiConnect[key].fieldGrouping = value;
+    },
+    'Expected auto, manual, or disabled.',
+  );
+}

src/config/resolve/anki-connect/initialize.ts

@@ -1,55 +1,8 @@
 import type { ResolveContext } from '../context';
-import { isObject } from '../shared';
-
-const LEGACY_KEYS = new Set([
-  'wordField',
-  'audioField',
-  'imageField',
-  'sentenceField',
-  'miscInfoField',
-  'miscInfoPattern',
-  'generateAudio',
-  'generateImage',
-  'imageType',
-  'imageFormat',
-  'imageQuality',
-  'imageMaxWidth',
-  'imageMaxHeight',
-  'animatedFps',
-  'animatedMaxWidth',
-  'animatedMaxHeight',
-  'animatedCrf',
-  'syncAnimatedImageToWordAudio',
-  'audioPadding',
-  'fallbackDuration',
-  'maxMediaDuration',
-  'overwriteAudio',
-  'overwriteImage',
-  'mediaInsertMode',
-  'highlightWord',
-  'notificationType',
-  'autoUpdateNewCards',
-]);
-
-export function initializeAnkiConnectResolution(
-  context: ResolveContext,
-  ankiConnect: Record<string, unknown>,
-): void {
-  const {
-    knownWords: _knownWordsConfigFromAnkiConnect,
-    nPlusOne: _nPlusOneConfigFromAnkiConnect,
-    ai: _ankiAiConfig,
-    ...ankiConnectWithoutKnownWordsOrNPlusOne
-  } = ankiConnect;
-  const ankiConnectWithoutLegacy = Object.fromEntries(
-    Object.entries(ankiConnectWithoutKnownWordsOrNPlusOne).filter(([key]) => !LEGACY_KEYS.has(key)),
-  );
 
+export function initializeAnkiConnectResolution(context: ResolveContext): void {
   context.resolved.ankiConnect = {
     ...context.resolved.ankiConnect,
-    ...(isObject(ankiConnectWithoutLegacy)
-      ? (ankiConnectWithoutLegacy as Partial<(typeof context.resolved)['ankiConnect']>)
-      : {}),
     fields: {
       ...context.resolved.ankiConnect.fields,
     },
@@ -73,15 +26,9 @@ export function initializeAnkiConnectResolution(
     },
     isKiku: {
       ...context.resolved.ankiConnect.isKiku,
-      ...(isObject(ankiConnect.isKiku)
-        ? (ankiConnect.isKiku as (typeof context.resolved)['ankiConnect']['isKiku'])
-        : {}),
     },
     isSenren: {
       ...context.resolved.ankiConnect.isSenren,
-      ...(isObject(ankiConnect.isSenren)
-        ? (ankiConnect.isSenren as (typeof context.resolved)['ankiConnect']['isSenren'])
-        : {}),
     },
     lapisKiku: {
       ...context.resolved.ankiConnect.lapisKiku,

src/config/resolve/anki-connect/kiku.ts

@@ -1,19 +1,9 @@
-import { DEFAULT_CONFIG } from '../../definitions';
 import type { ResolveContext } from '../context';
+import { applyFieldGroupingConfigResolution } from './field-grouping-config';
 
-export function applyAnkiKikuResolution(context: ResolveContext): void {
-  if (
-    context.resolved.ankiConnect.isKiku.fieldGrouping !== 'auto' &&
-    context.resolved.ankiConnect.isKiku.fieldGrouping !== 'manual' &&
-    context.resolved.ankiConnect.isKiku.fieldGrouping !== 'disabled'
-  ) {
-    context.warn(
-      'ankiConnect.isKiku.fieldGrouping',
-      context.resolved.ankiConnect.isKiku.fieldGrouping,
-      DEFAULT_CONFIG.ankiConnect.isKiku.fieldGrouping,
-      'Expected auto, manual, or disabled.',
-    );
-    context.resolved.ankiConnect.isKiku.fieldGrouping =
-      DEFAULT_CONFIG.ankiConnect.isKiku.fieldGrouping;
-  }
+export function applyAnkiKikuResolution(
+  context: ResolveContext,
+  ankiConnect: Record<string, unknown>,
+): void {
+  applyFieldGroupingConfigResolution(context, ankiConnect, 'isKiku');
 }

src/config/resolve/anki-connect/modern-fields.ts

@@ -7,7 +7,15 @@ export function applyModernFieldsResolution(
   context: ResolveContext,
   fields: Record<string, unknown>,
 ): void {
-  for (const key of ['word', 'audio', 'image', 'sentence', 'miscInfo', 'translation'] as const) {
+  for (const key of [
+    'word',
+    'audio',
+    'wordAudio',
+    'image',
+    'sentence',
+    'miscInfo',
+    'translation',
+  ] as const) {
     applyModernValue(
       context,
       fields,

src/config/resolve/anki-connect/modern.ts

@@ -1,6 +1,7 @@
 import type { ResolveContext } from '../context';
 import { isObject } from '../shared';
 import { applyAiResolution } from './ai';
+import { applyAnkiBaseResolution } from './base';
 import { applyLapisResolution } from './lapis';
 import { applyModernBehaviorResolution } from './modern-behavior';
 import { applyModernFieldsResolution } from './modern-fields';
@@ -18,6 +19,7 @@ export function applyAnkiModernResolution(
   const fields = isObject(ankiConnect.fields) ? ankiConnect.fields : {};
   const metadata = isObject(ankiConnect.metadata) ? ankiConnect.metadata : {};
 
+  applyAnkiBaseResolution(context, ankiConnect);
   applyModernFieldsResolution(context, fields);
   applyModernMediaResolution(context, media);
   applyModernBehaviorResolution(context, behavior);

src/config/resolve/anki-connect/senren.ts

@@ -1,21 +1,11 @@
-import { DEFAULT_CONFIG } from '../../definitions';
 import type { ResolveContext } from '../context';
+import { applyFieldGroupingConfigResolution } from './field-grouping-config';
 
-export function applyAnkiSenrenResolution(context: ResolveContext): void {
-  if (
-    context.resolved.ankiConnect.isSenren.fieldGrouping !== 'auto' &&
-    context.resolved.ankiConnect.isSenren.fieldGrouping !== 'manual' &&
-    context.resolved.ankiConnect.isSenren.fieldGrouping !== 'disabled'
-  ) {
-    context.warn(
-      'ankiConnect.isSenren.fieldGrouping',
-      context.resolved.ankiConnect.isSenren.fieldGrouping,
-      DEFAULT_CONFIG.ankiConnect.isSenren.fieldGrouping,
-      'Expected auto, manual, or disabled.',
-    );
-    context.resolved.ankiConnect.isSenren.fieldGrouping =
-      DEFAULT_CONFIG.ankiConnect.isSenren.fieldGrouping;
-  }
+export function applyAnkiSenrenResolution(
+  context: ResolveContext,
+  ankiConnect: Record<string, unknown>,
+): void {
+  applyFieldGroupingConfigResolution(context, ankiConnect, 'isSenren');
 
   // Kiku and Senren field grouping write incompatible markup into the same note
   // fields, so only one may be active; Kiku wins to preserve pre-existing setups.

src/config/resolve/integrations.ts

@@ -59,6 +59,19 @@ export function applyIntegrationConfig(context: ResolveContext): void {
     warn('ai', src.ai, resolved.ai, 'Expected object.');
   }
 
+  if (isObject(src.tmdb)) {
+    for (const key of ['apiKey', 'apiKeyCommand'] as const) {
+      const value = asString(src.tmdb[key]);
+      if (value !== undefined) {
+        resolved.tmdb[key] = value;
+      } else if (src.tmdb[key] !== undefined) {
+        warn(`tmdb.${key}`, src.tmdb[key], resolved.tmdb[key], 'Expected string.');
+      }
+    }
+  } else if (src.tmdb !== undefined) {
+    warn('tmdb', src.tmdb, resolved.tmdb, 'Expected object.');
+  }
+
   if (isObject(src.anilist)) {
     const enabled = asBoolean(src.anilist.enabled);
     if (enabled !== undefined) {

src/config/settings/registry.test.ts

@@ -290,6 +290,9 @@ test('settings registry routes playback-related integrations into integrations',
   assert.equal(field('subsync.replace').section, 'Subtitle Sync');
   assert.equal(field('tsukihime.apiBaseUrl').category, 'integrations');
   assert.equal(field('tsukihime.apiBaseUrl').section, 'TsukiHime');
+  assert.equal(field('tmdb.apiKey').category, 'integrations');
+  assert.equal(field('tmdb.apiKey').section, 'TMDB');
+  assert.equal(field('tmdb.apiKey').secret, true);
 });
 
 test('settings registry puts feature toggles first, then other toggles alphabetically', () => {
@@ -379,6 +382,7 @@ test('settings registry marks safe live config paths as hot-reloadable', () => {
     'ankiConnect.nPlusOne.minSentenceWords',
     'ankiConnect.fields.word',
     'ankiConnect.fields.audio',
+    'ankiConnect.fields.wordAudio',
     'ankiConnect.fields.image',
     'ankiConnect.fields.sentence',
     'ankiConnect.fields.miscInfo',

src/config/settings/registry.ts

@@ -1,9 +1,9 @@
+import { getConfigHotReloadField } from '../hot-reload';
 import type { ResolvedConfig } from '../../types/config';
 import type {
   ConfigSettingsCategory,
   ConfigSettingsControl,
   ConfigSettingsField,
-  ConfigSettingsRestartBehavior,
 } from '../../types/settings';
 import { CONFIG_OPTION_REGISTRY, DEFAULT_CONFIG } from '../definitions';
 import {
@@ -93,7 +93,12 @@ const JSON_OBJECT_FIELDS = new Set([
   'subtitleSidebar.css',
 ]);
 
-export const SECRET_PATHS = new Set(['ai.apiKey', 'jimaku.apiKey', 'anilist.accessToken']);
+export const SECRET_PATHS = new Set([
+  'ai.apiKey',
+  'jimaku.apiKey',
+  'tmdb.apiKey',
+  'anilist.accessToken',
+]);
 
 const COLOR_SUFFIXES = new Set(['Color', 'color', 'backgroundColor', 'singleColor']);
 const SUBTITLE_CSS_MANAGED_CONFIG_PATHS = new Set([
@@ -136,6 +141,7 @@ const SECTION_ORDER = new Map<string, number>(
     'AnkiConnect Proxy',
     'Aniyomi',
     'Jimaku',
+    'TMDB',
     'Subtitle Sync',
     'MPV Keybindings',
     'Overlay Shortcuts',
@@ -329,6 +335,7 @@ function humanizePath(path: string): string {
     .replace(/\bmpv\b/i, 'mpv')
     .replace(/\byomitan\b/i, 'Yomitan')
     .replace(/\bjimaku\b/i, 'Jimaku')
+    .replace(/\btmdb\b/i, 'TMDB')
     .replace(/\banilist\b/i, 'AniList')
     .replace(/\banki\b/i, 'Anki');
   return spaced.charAt(0).toUpperCase() + spaced.slice(1);
@@ -444,7 +451,7 @@ function categoryAndSection(path: string): { category: ConfigSettingsCategory; s
   if (path.startsWith('mpv.') || path.startsWith('youtube.')) {
     return { category: 'behavior', section: topSection(path) };
   }
-  if (path.startsWith('jimaku.') || path.startsWith('tsukihime.')) {
+  if (path.startsWith('jimaku.') || path.startsWith('tsukihime.') || path.startsWith('tmdb.')) {
     return { category: 'integrations', section: topSection(path) };
   }
   if (path.startsWith('anime.')) {
@@ -515,6 +522,7 @@ function topSection(path: string): string {
     subsync: 'Subtitle Sync',
     texthooker: 'Texthooker',
     tsukihime: 'TsukiHime',
+    tmdb: 'TMDB',
     updates: 'Updates',
     websocket: 'WebSocket server',
     yomitan: 'Yomitan',
@@ -698,54 +706,6 @@ function compareFields(a: ConfigSettingsField, b: ConfigSettingsField): number {
   return a.configPath.localeCompare(b.configPath);
 }
 
-function restartBehaviorForPath(path: string): ConfigSettingsRestartBehavior {
-  if (
-    path === 'keybindings' ||
-    pathStartsWith(path, 'shortcuts') ||
-    pathStartsWith(path, 'subtitleStyle') ||
-    pathStartsWith(path, 'subtitleSidebar') ||
-    path === 'secondarySub.defaultMode' ||
-    path === 'ankiConnect.deck' ||
-    path === 'ankiConnect.ai.enabled' ||
-    path === 'ankiConnect.media.normalizeAudio' ||
-    path === 'ankiConnect.media.mirrorMpvVolume' ||
-    path === 'ankiConnect.media.reviewTiming' ||
-    path === 'ankiConnect.behavior.autoUpdateNewCards' ||
-    path === 'ankiConnect.knownWords.highlightEnabled' ||
-    path === 'ankiConnect.knownWords.refreshMinutes' ||
-    path === 'ankiConnect.knownWords.addMinedWordsImmediately' ||
-    path === 'ankiConnect.knownWords.matchMode' ||
-    path === 'ankiConnect.knownWords.decks' ||
-    path === 'ankiConnect.nPlusOne.enabled' ||
-    path === 'ankiConnect.nPlusOne.minSentenceWords' ||
-    path === 'ankiConnect.fields.word' ||
-    path === 'ankiConnect.fields.audio' ||
-    path === 'ankiConnect.fields.image' ||
-    path === 'ankiConnect.fields.sentence' ||
-    path === 'ankiConnect.fields.miscInfo' ||
-    path === 'ankiConnect.isLapis.sentenceCardModel' ||
-    path === 'ankiConnect.isKiku.fieldGrouping' ||
-    path === 'ankiConnect.isSenren.fieldGrouping' ||
-    path === 'ankiConnect.lapisKiku.wordCardKind' ||
-    path === 'mpv.aniskipEnabled' ||
-    path === 'mpv.aniskipButtonKey' ||
-    path === 'stats.toggleKey' ||
-    path === 'stats.markWatchedKey' ||
-    path === 'logging.level' ||
-    path === 'logging.rotation' ||
-    pathStartsWith(path, 'logging.files') ||
-    pathStartsWith(path, 'notifications') ||
-    path === 'anime.autoOpenJimaku' ||
-    path === 'youtube.primarySubLanguages' ||
-    pathStartsWith(path, 'jimaku') ||
-    pathStartsWith(path, 'subsync') ||
-    pathStartsWith(path, 'subtitleGeneration')
-  ) {
-    return 'hot-reload';
-  }
-  return 'restart';
-}
-
 function fieldForLeaf(leaf: Leaf): ConfigSettingsField {
   const option = OPTION_BY_PATH.get(leaf.path);
   const { category, section } = categoryAndSection(leaf.path);
@@ -764,7 +724,7 @@ function fieldForLeaf(leaf: Leaf): ConfigSettingsField {
       ? { enumValues: option.settingsEnumValues ?? option.enumValues }
       : {}),
     ...(option?.enumLabels ? { enumLabels: option.enumLabels } : {}),
-    restartBehavior: restartBehaviorForPath(leaf.path),
+    restartBehavior: getConfigHotReloadField(leaf.path) ? 'hot-reload' : 'restart',
     advanced:
       leaf.path.startsWith('controller.') ||
       leaf.path.startsWith('immersionTracking.retention.') ||

src/core/services/__tests__/stats-server.test.ts

@@ -5,8 +5,13 @@ import http from 'node:http';
 import os from 'node:os';
 import path from 'node:path';
 import type { AddressInfo } from 'node:net';
-import { createStatsApp, startStatsServer } from '../stats-server.js';
+import {
+  createStatsApp,
+  startNodeHttpServer,
+  startStatsServerWithRuntime,
+} from '../stats-server.js';
 import type { ImmersionTrackerService } from '../immersion-tracker-service.js';
+import { INCOMPATIBLE_PROVIDER_MERGE_MESSAGE } from '../immersion-tracker/anime-merge.js';
 import {
   clearRetimedSecondarySubtitleCache,
   resolveRetimedSecondarySubtitleTextFromSidecar,
@@ -307,6 +312,7 @@ function createMockTracker(
     getKanjiOccurrences: async () => OCCURRENCES,
     getAnimeLibrary: async () => ANIME_LIBRARY,
     getAnimeDetail: async (animeId: number) => (animeId === 1 ? ANIME_DETAIL : null),
+    hasAnime: async (animeId: number) => animeId === 1,
     getAnimeEpisodes: async () => ANIME_EPISODES,
     getAnimeAnilistEntries: async () => [],
     getAnimeWords: async () => ANIME_WORDS,
@@ -440,6 +446,73 @@ async function withFakeAnkiConnect<T>(
 }
 
 describe('stats server API routes', () => {
+  it('rejects untrusted mutation requests before merging anime', async () => {
+    let merges = 0;
+    const app = createStatsApp(
+      createMockTracker({
+        mergeAnime: async () => {
+          merges += 1;
+          return { survivingAnimeId: 1, mergedAnimeIds: [2], movedVideos: 1 };
+        },
+      }),
+    );
+    const rejectedHeaders: Record<string, string>[] = [
+      { Origin: 'https://attacker.example', 'Content-Type': 'text/plain' },
+      { Origin: 'https://attacker.example', 'Content-Type': 'application/json' },
+      { Origin: 'null', 'Content-Type': 'application/json' },
+      { Origin: 'http://localhost:4321', 'Content-Type': 'application/json' },
+      { Origin: 'http://localhost/', 'Content-Type': 'application/json' },
+      { 'Sec-Fetch-Site': 'cross-site', 'Content-Type': 'application/json' },
+      { Host: 'attacker.example', 'Content-Type': 'application/json' },
+    ];
+    for (const headers of rejectedHeaders) {
+      const response = await app.request('/api/stats/anime/1/merge', {
+        method: 'POST',
+        headers,
+        body: JSON.stringify({ sourceAnimeIds: [2] }),
+      });
+      assert.equal(response.status, 403, JSON.stringify(headers));
+    }
+    assert.equal(merges, 0);
+    for (const origin of [undefined, 'http://localhost']) {
+      const headers = new Headers({ 'Content-Type': 'application/json; charset=utf-8' });
+      if (origin) headers.set('Origin', origin);
+      const response = await app.request('/api/stats/anime/1/merge', {
+        method: 'POST',
+        headers,
+        body: JSON.stringify({ sourceAnimeIds: [2] }),
+      });
+      assert.equal(response.status, 200);
+    }
+    assert.equal(merges, 2);
+  });
+
+  it('requires JSON for mutation bodies and preserves bodyless deletion', async () => {
+    let deletions = 0;
+    const app = createStatsApp(
+      createMockTracker({
+        deleteSession: async () => {
+          deletions += 1;
+        },
+      }),
+    );
+    const invalid = await app.request('/api/stats/sessions/1', {
+      method: 'DELETE',
+      body: '{}',
+    });
+    assert.equal(invalid.status, 415);
+    assert.equal(deletions, 0);
+    const valid = await app.request('/api/stats/sessions/1', { method: 'DELETE' });
+    assert.equal(valid.status, 200);
+    assert.equal(deletions, 1);
+    const rebound = await app.request('http://attacker.example/api/stats/sessions/1', {
+      method: 'DELETE',
+      headers: { Origin: 'http://attacker.example' },
+    });
+    assert.equal(rebound.status, 403);
+    assert.equal(deletions, 1);
+  });
+
   it('GET /api/stats/overview returns overview data', async () => {
     const app = createStatsApp(createMockTracker());
     const res = await app.request('/api/stats/overview');
@@ -1004,6 +1077,23 @@ describe('stats server API routes', () => {
     assert.equal(seenLimit, 500);
   });
 
+  it('GET /api/stats/vocabulary floors fractional pagination limits', async () => {
+    let seenLimit = 0;
+    const app = createStatsApp(
+      createMockTracker({
+        getVocabularyStats: async (limit?: number) => {
+          seenLimit = limit ?? 0;
+          return VOCABULARY_STATS;
+        },
+      }),
+    );
+
+    const res = await app.request('/api/stats/vocabulary?limit=12.9');
+
+    assert.equal(res.status, 200);
+    assert.equal(seenLimit, 12);
+  });
+
   it('GET /api/stats/vocabulary passes excludePos to tracker', async () => {
     let seenArgs: unknown[] = [];
     const app = createStatsApp(
@@ -1132,7 +1222,7 @@ describe('stats server API routes', () => {
       body: JSON.stringify({ dryRun: false, lookbackDays: null }),
     });
 
-    assert.equal(res.status, 415);
+    assert.equal(res.status, 403);
     assert.equal(cleanupCalls, 0);
   });
 
@@ -1351,7 +1441,7 @@ describe('stats server API routes', () => {
       }),
     );
 
-    for (const anilistId of [-1, 0, 1.5, '12', true, undefined]) {
+    for (const anilistId of [-1, 0, 1.5, 9_007_199_254_740_992, '12', true, undefined]) {
       const res = await app.request('/api/stats/anime/1/anilist', {
         method: 'PATCH',
         headers: { 'Content-Type': 'application/json' },
@@ -1449,6 +1539,74 @@ describe('stats server API routes', () => {
     assert.equal(res.status, 404);
   });
 
+  it('resource routes reject fractional ids before calling dependencies', async () => {
+    const dependencyCalls: string[] = [];
+    const originalFetch = globalThis.fetch;
+    globalThis.fetch = async () => {
+      dependencyCalls.push('fetch');
+      return new Response('{}', { status: 200 });
+    };
+
+    try {
+      const app = createStatsApp(
+        createMockTracker({
+          getWordDetail: async () => {
+            dependencyCalls.push('getWordDetail');
+            return null;
+          },
+          getSessionEvents: async () => {
+            dependencyCalls.push('getSessionEvents');
+            return [];
+          },
+          getEpisodeSessions: async () => {
+            dependencyCalls.push('getEpisodeSessions');
+            return [];
+          },
+          getAnimeCoverArt: async () => {
+            dependencyCalls.push('getAnimeCoverArt');
+            return null;
+          },
+          ensureAnimeCoverArt: async () => {
+            dependencyCalls.push('ensureAnimeCoverArt');
+            return false;
+          },
+          setVideoWatched: async () => {
+            dependencyCalls.push('setVideoWatched');
+          },
+          reassignAnimeAnilist: async () => {
+            dependencyCalls.push('reassignAnimeAnilist');
+          },
+        }),
+      );
+
+      const responses = await Promise.all([
+        app.request('/api/stats/vocabulary/1.9/detail'),
+        app.request('/api/stats/sessions/1.9/events'),
+        app.request('/api/stats/episode/1.9/detail'),
+        app.request('/api/stats/anime/1.9/cover'),
+        app.request('/api/stats/media/1.9/watched', {
+          method: 'PATCH',
+          headers: { 'Content-Type': 'application/json' },
+          body: '{"watched":true}',
+        }),
+        app.request('/api/stats/anime/1.9/anilist', {
+          method: 'PATCH',
+          headers: { 'Content-Type': 'application/json' },
+          body: '{"anilistId":21858}',
+        }),
+        app.request('/api/stats/anki/browse?noteId=1.9', { method: 'POST' }),
+      ]);
+
+      assert.deepEqual(
+        responses.map((response) => response.status),
+        [400, 400, 400, 400, 400, 400, 400],
+      );
+      assert.deepEqual(dependencyCalls, []);
+    } finally {
+      globalThis.fetch = originalFetch;
+    }
+  });
+
   it('POST /api/stats/covers batches stored cover art and backfills missing anime art in the background', async () => {
     let ensureCoverArtCalls = 0;
     const ensureAnimeCoverArtCalls: number[] = [];
@@ -1505,6 +1663,58 @@ describe('stats server API routes', () => {
     assert.deepEqual(ensureAnimeCoverArtCalls, [99999]);
   });
 
+  it('JSON id lists reject malformed members before side effects', async () => {
+    const dependencyCalls: string[] = [];
+    const originalFetch = globalThis.fetch;
+    globalThis.fetch = async () => {
+      dependencyCalls.push('fetch');
+      return new Response('{}', { status: 200 });
+    };
+
+    try {
+      const app = createStatsApp(
+        createMockTracker({
+          deleteSessions: async () => {
+            dependencyCalls.push('deleteSessions');
+          },
+          mergeAnime: async () => {
+            dependencyCalls.push('mergeAnime');
+            return { survivingAnimeId: 7, mergedAnimeIds: [], movedVideos: 0 };
+          },
+          getAnimeCoverArt: async () => {
+            dependencyCalls.push('getAnimeCoverArt');
+            return null;
+          },
+          ensureAnimeCoverArt: async () => {
+            dependencyCalls.push('ensureAnimeCoverArt');
+            return false;
+          },
+        }),
+      );
+      const request = async (path: string, body: string, method = 'POST'): Promise<Response> =>
+        await app.request(path, {
+          method,
+          headers: { 'Content-Type': 'application/json' },
+          body,
+        });
+
+      const responses = await Promise.all([
+        request('/api/stats/sessions', '{"sessionIds":[4,1.9,7]}', 'DELETE'),
+        request('/api/stats/anime/7/merge', '{"sourceAnimeIds":[8,"9"]}'),
+        request('/api/stats/covers', '{"animeIds":[1,1.9]}'),
+        request('/api/stats/anki/notesInfo', '{"noteIds":[1,1.9]}'),
+      ]);
+
+      assert.deepEqual(
+        responses.map((response) => response.status),
+        [400, 400, 400, 400],
+      );
+      assert.deepEqual(dependencyCalls, []);
+    } finally {
+      globalThis.fetch = originalFetch;
+    }
+  });
+
   it('POST /api/stats/covers limits concurrent missing anime cover backfills', async () => {
     let activeBackfills = 0;
     let maxActiveBackfills = 0;
@@ -1762,6 +1972,60 @@ describe('stats server API routes', () => {
     });
   });
 
+  it('POST /api/stats/mine-card treats a zero media duration cap as unlimited', async () => {
+    await withTempDir(async (dir) => {
+      const sourcePath = path.join(dir, 'episode.mkv');
+      fs.writeFileSync(sourcePath, 'fake media');
+      const audioRanges: Array<{ start: number; end: number; padding: number | undefined }> = [];
+      const scenarios = [
+        { maxMediaDuration: 0, expectedEnd: 12 },
+        { maxMediaDuration: 1, expectedEnd: 11 },
+      ];
+
+      for (const scenario of scenarios) {
+        const app = createStatsApp(createMockTracker(), {
+          addYomitanNote: async () => null,
+          createMediaGenerator: () => ({
+            generateAudio: async (_path, start, end, padding) => {
+              audioRanges.push({ start, end, padding });
+              return Buffer.from('audio');
+            },
+            generateScreenshot: async () => null,
+            generateAnimatedImage: async () => null,
+          }),
+          ankiConnectConfig: {
+            deck: 'Mining',
+            media: {
+              generateAudio: true,
+              generateImage: false,
+              audioPadding: 0.25,
+              maxMediaDuration: scenario.maxMediaDuration,
+            },
+          },
+        });
+
+        const res = await app.request('/api/stats/mine-card?mode=word', {
+          method: 'POST',
+          headers: { 'Content-Type': 'application/json' },
+          body: JSON.stringify({
+            sourcePath,
+            startMs: 10_000,
+            endMs: 12_000,
+            sentence: '猫を見た',
+            word: '猫',
+          }),
+        });
+
+        assert.equal(res.status, 502);
+        assert.deepEqual(audioRanges.at(-1), {
+          start: 10,
+          end: scenario.expectedEnd,
+          padding: 0.25,
+        });
+      }
+    });
+  });
+
   it('POST /api/stats/mine-card requires a non-empty word in word mode', async () => {
     await withTempDir(async (dir) => {
       const sourcePath = path.join(dir, 'episode.mkv');
@@ -3262,6 +3526,46 @@ Aligned English subtitle
     assert.equal(deleteCalls, 0);
   });
 
+  it('DELETE /api/stats/sessions rejects a partly invalid id list without deleting', async () => {
+    let deleteCalls = 0;
+    const app = createStatsApp(
+      createMockTracker({
+        deleteSessions: async () => {
+          deleteCalls += 1;
+        },
+      }),
+    );
+
+    const res = await app.request('/api/stats/sessions', {
+      method: 'DELETE',
+      headers: { 'Content-Type': 'application/json' },
+      body: '{"sessionIds":[4,1.9,7]}',
+    });
+
+    assert.equal(res.status, 400);
+    assert.equal(deleteCalls, 0);
+  });
+
+  it('DELETE /api/stats/sessions deduplicates valid ids', async () => {
+    let deletedSessionIds: number[] = [];
+    const app = createStatsApp(
+      createMockTracker({
+        deleteSessions: async (sessionIds: number[]) => {
+          deletedSessionIds = sessionIds;
+        },
+      }),
+    );
+
+    const res = await app.request('/api/stats/sessions', {
+      method: 'DELETE',
+      headers: { 'Content-Type': 'application/json' },
+      body: '{"sessionIds":[4,4,7]}',
+    });
+
+    assert.equal(res.status, 200);
+    assert.deepEqual(deletedSessionIds, [4, 7]);
+  });
+
   it('DELETE /api/stats/anime/:animeId deletes the whole library entry', async () => {
     let deletedAnimeId: number | null = null;
     const app = createStatsApp(
@@ -3295,6 +3599,33 @@ Aligned English subtitle
     assert.equal(deleteCalls, 0);
   });
 
+  it('DELETE /api/stats/anime/:animeId rejects malformed anime ids before deleting', async () => {
+    let deletedAnimeId: number | null = null;
+    const app = createStatsApp(
+      createMockTracker({
+        deleteAnime: async (animeId: number) => {
+          deletedAnimeId = animeId;
+        },
+      }),
+    );
+
+    for (const animeId of [
+      '1.9',
+      '1.0',
+      '1e2',
+      '9007199254740992',
+      '1%0A',
+      '%201',
+      '01',
+      '+1',
+      '0x1',
+    ]) {
+      const res = await app.request(`/api/stats/anime/${animeId}`, { method: 'DELETE' });
+      assert.equal(res.status, 400, `accepted malformed anime id: ${animeId}`);
+    }
+    assert.equal(deletedAnimeId, null);
+  });
+
   it('POST /api/stats/anime/:animeId/merge folds the given entries into the target', async () => {
     let merged: { targetAnimeId: number; sourceAnimeIds: number[] } | null = null;
     const app = createStatsApp(
@@ -3400,6 +3731,25 @@ Aligned English subtitle
     assert.equal(res.status, 404);
   });
 
+  it('POST /api/stats/anime/:animeId/merge rejects mixed AniList and TMDB entries as 409', async () => {
+    const app = createStatsApp(
+      createMockTracker({
+        mergeAnime: async () => {
+          throw new Error(INCOMPATIBLE_PROVIDER_MERGE_MESSAGE);
+        },
+      } as Partial<ImmersionTrackerService>),
+    );
+
+    const res = await app.request('/api/stats/anime/7/merge', {
+      method: 'POST',
+      headers: { 'Content-Type': 'application/json' },
+      body: '{"sourceAnimeIds":[8]}',
+    });
+
+    assert.equal(res.status, 409);
+    assert.deepEqual(await res.json(), { error: INCOMPATIBLE_PROVIDER_MERGE_MESSAGE });
+  });
+
   it('PATCH /api/stats/media/:videoId/anime reports an unknown target as 404', async () => {
     const app = createStatsApp(
       createMockTracker({
@@ -3737,102 +4087,237 @@ Aligned English subtitle
     assert.equal(ensureCalls, 1);
   });
 
-  it('starts the stats server with Bun.serve', () => {
-    type BunRuntime = {
-      Bun: {
-        serve: (options: { fetch: unknown; port: number; hostname: string }) => {
-          stop: () => void;
-        };
-      };
-    };
-
-    const bun = globalThis as typeof globalThis & BunRuntime;
-    const originalServe = bun.Bun.serve;
-    let servedWith: { fetch: unknown; port: number; hostname: string } | null = null;
+  it('starts and stops the stats server with Bun.serve', async () => {
+    const servedOptions: Array<{ fetch: unknown; port: number; hostname: string }> = [];
     let stopCalls = 0;
-
-    bun.Bun.serve = (options: { fetch: unknown; port: number; hostname: string }) => {
-      servedWith = options;
-      return {
-        stop: () => {
-          stopCalls += 1;
-        },
-      };
-    };
-
-    try {
-      const server = startStatsServer({
+    const server = await startStatsServerWithRuntime(
+      {
         port: 3210,
         staticDir: fs.mkdtempSync(path.join(os.tmpdir(), 'subminer-stats-server-start-')),
         tracker: createMockTracker(),
-      });
+      },
+      {
+        bunServe: (options) => {
+          servedOptions.push(options);
+          return {
+            stop: () => {
+              stopCalls += 1;
+            },
+          };
+        },
+      },
+    );
 
-      if (servedWith === null) {
-        throw new Error('expected Bun.serve to be called');
-      }
-
-      const servedOptions = servedWith as {
-        fetch: unknown;
-        port: number;
-        hostname: string;
-      };
-      assert.equal(servedOptions.port, 3210);
-      assert.equal(servedOptions.hostname, '127.0.0.1');
-      assert.equal(typeof servedOptions.fetch, 'function');
-
-      server.close();
-      assert.equal(stopCalls, 1);
-    } finally {
-      bun.Bun.serve = originalServe;
+    const servedWith = servedOptions[0];
+    if (!servedWith) {
+      throw new Error('expected Bun.serve to be called');
     }
+
+    assert.equal(servedWith.port, 3210);
+    assert.equal(servedWith.hostname, '127.0.0.1');
+    assert.equal(typeof servedWith.fetch, 'function');
+
+    await Promise.all([server.close(), server.close()]);
+    assert.equal(stopCalls, 1);
   });
 
-  it('falls back to node:http when Bun.serve is unavailable', () => {
-    type BunRuntime = {
-      Bun: {
-        serve?: (options: { fetch: unknown; port: number; hostname: string }) => {
-          stop: () => void;
-        };
-      };
-    };
-
-    const bun = globalThis as typeof globalThis & BunRuntime;
-    const originalServe = bun.Bun.serve;
-    const originalCreateServer = http.createServer;
-    let listenedWith: { port: number; hostname: string } | null = null;
+  it('waits for node:http listening and converts startup errors into rejections', async () => {
+    const app = createStatsApp(createMockTracker());
+    const listeningServer = http.createServer();
     let closeCalls = 0;
-    bun.Bun.serve = undefined;
-    (
-      http as typeof http & {
-        createServer: typeof http.createServer;
-      }
-    ).createServer = (() =>
-      ({
-        listen: (port: number, hostname: string) => {
-          listenedWith = { port, hostname };
-        },
-        close: () => {
+    Object.defineProperties(listeningServer, {
+      listen: {
+        value: () => listeningServer,
+      },
+      close: {
+        value: (callback?: (error?: Error) => void) => {
           closeCalls += 1;
+          callback?.();
+          return listeningServer;
         },
-      }) as unknown as ReturnType<typeof http.createServer>) as typeof http.createServer;
+      },
+    });
+
+    let startupSettled = false;
+    const startup = startNodeHttpServer(
+      app,
+      {
+        port: 3210,
+        staticDir: fs.mkdtempSync(path.join(os.tmpdir(), 'subminer-stats-server-node-events-')),
+        tracker: createMockTracker(),
+      },
+      () => listeningServer,
+    );
+    void startup.finally(() => {
+      startupSettled = true;
+    });
+    await Promise.resolve();
+    assert.equal(startupSettled, false);
+
+    listeningServer.emit('listening');
+    const handle = await startup;
+    await Promise.all([handle.close(), handle.close()]);
+    assert.equal(closeCalls, 1);
+
+    const failingServer = http.createServer();
+    Object.defineProperty(failingServer, 'listen', {
+      value: () => failingServer,
+    });
+    const failedStartup = startNodeHttpServer(
+      app,
+      {
+        port: 3210,
+        staticDir: fs.mkdtempSync(path.join(os.tmpdir(), 'subminer-stats-server-node-error-')),
+        tracker: createMockTracker(),
+      },
+      () => failingServer,
+    );
+    failingServer.emit('error', Object.assign(new Error('address in use'), { code: 'EADDRINUSE' }));
+    await assert.rejects(
+      failedStartup,
+      (error: NodeJS.ErrnoException) => error.code === 'EADDRINUSE',
+    );
+  });
+
+  it('starts, rejects address conflicts, and stops through real node:http sockets', async () => {
+    const app = createStatsApp(createMockTracker());
+    const server = await startNodeHttpServer(app, {
+      port: 0,
+      staticDir: fs.mkdtempSync(path.join(os.tmpdir(), 'subminer-stats-server-node-')),
+      tracker: createMockTracker(),
+    });
+    await Promise.all([server.close(), server.close()]);
+
+    const blocker = http.createServer();
+    await new Promise<void>((resolve, reject) => {
+      blocker.once('error', reject);
+      blocker.listen(0, '127.0.0.1', resolve);
+    });
+    const address = blocker.address();
+    if (!address || typeof address === 'string') {
+      throw new Error('expected blocker to listen on a TCP port');
+    }
 
     try {
-      const server = startStatsServer({
-        port: 0,
-        staticDir: fs.mkdtempSync(path.join(os.tmpdir(), 'subminer-stats-server-node-')),
-        tracker: createMockTracker(),
-      });
-
-      assert.deepEqual(listenedWith, { port: 0, hostname: '127.0.0.1' });
-      server.close();
-      assert.equal(closeCalls, 1);
+      await assert.rejects(
+        startNodeHttpServer(app, {
+          port: address.port,
+          staticDir: fs.mkdtempSync(path.join(os.tmpdir(), 'subminer-stats-server-node-error-')),
+          tracker: createMockTracker(),
+        }),
+        (error: NodeJS.ErrnoException) => error.code === 'EADDRINUSE',
+      );
     } finally {
-      bun.Bun.serve = originalServe;
-      (
-        http as typeof http & {
-          createServer: typeof http.createServer;
-        }
-      ).createServer = originalCreateServer;
+      await new Promise<void>((resolve, reject) => {
+        blocker.close((error) => {
+          if (error) reject(error);
+          else resolve();
+        });
+      });
     }
   });
+
+  it('enforces request safety through node:http without rejecting bodyless DELETEs', async () => {
+    await withTempDir(async (staticDir) => {
+      let deletions = 0;
+      const tracker = createMockTracker({
+        deleteSession: async () => {
+          deletions += 1;
+        },
+      });
+      const listener = http.createServer();
+      const server = await startNodeHttpServer(
+        createStatsApp(tracker),
+        { port: 0, staticDir, tracker },
+        (handler) => {
+          listener.on('request', handler);
+          return listener;
+        },
+      );
+      try {
+        const address = listener.address();
+        assert.ok(address && typeof address !== 'string');
+        const origin = `http://127.0.0.1:${address.port}`;
+        const url = `${origin}/api/stats/sessions/1`;
+        for (const headers of [undefined, { 'Content-Length': '0' }]) {
+          const response = await fetch(url, { method: 'DELETE', headers });
+          assert.equal(response.status, 200);
+          await response.arrayBuffer();
+        }
+        assert.equal(deletions, 2);
+        for (const headers of [
+          new Headers({ Origin: 'https://attacker.example' }),
+          new Headers({ Origin: 'null' }),
+          new Headers({ Host: 'attacker.example' }),
+          new Headers({ 'Sec-Fetch-Site': 'same-site' }),
+        ]) {
+          const response = await fetch(url, { method: 'DELETE', headers });
+          assert.equal(response.status, 403, JSON.stringify(headers));
+          await response.arrayBuffer();
+        }
+        const invalid = await fetch(url, { method: 'DELETE', body: '{}' });
+        assert.equal(invalid.status, 415);
+        await invalid.arrayBuffer();
+        assert.equal(deletions, 2);
+        const valid = await fetch(url, {
+          method: 'DELETE',
+          headers: { Origin: origin, 'Content-Type': 'application/json' },
+          body: '{}',
+        });
+        assert.equal(valid.status, 200);
+        await valid.arrayBuffer();
+        assert.equal(deletions, 3);
+      } finally {
+        await server.close();
+      }
+    });
+  });
+});
+
+it('TMDB reassignment returns 404 for a missing library entry before fetching details', async () => {
+  const assignments: number[] = [];
+  let fetches = 0;
+  const app = createStatsApp(
+    createMockTracker({
+      reassignAnimeTmdb: async (animeId: number) => {
+        assignments.push(animeId);
+        return { animeId, mergedAnimeIds: [] };
+      },
+    }),
+    {
+      tmdbClient: {
+        search: async () => [],
+        getDetails: async () => {
+          fetches += 1;
+          return {
+            tmdbId: 12,
+            tmdbType: 'tv',
+            titleEnglish: 'Drama',
+            titleNative: null,
+            description: null,
+            posterUrl: null,
+            episodesTotal: 10,
+            year: null,
+            originalLanguage: 'ja',
+            isAnimation: false,
+            allTitles: ['Drama'],
+          };
+        },
+      },
+    },
+  );
+  const request = (animeId: number) =>
+    app.request(`/api/stats/anime/${animeId}/tmdb`, {
+      method: 'PATCH',
+      headers: { 'Content-Type': 'application/json' },
+      body: JSON.stringify({ tmdbId: 12, tmdbType: 'tv' }),
+    });
+  assert.equal((await request(99999)).status, 404);
+  assert.equal(fetches, 0);
+  assert.deepEqual(assignments, []);
+  const response = await request(1);
+  assert.equal(response.status, 200);
+  assert.deepEqual(await response.json(), { ok: true });
+  assert.equal(fetches, 1);
+  assert.deepEqual(assignments, [1]);
 });

src/core/services/anilist/anilist-update-queue.test.ts

@@ -27,6 +27,83 @@ function createLogger() {
   };
 }
 
+test('anilist retry queue migrates stream keys and rejects URL-derived searches', () => {
+  const queueFile = createTempQueueFile();
+  const loggerState = createLogger();
+  const key = 'https://example.com/Videos/item/stream?api_key=test-secret::2';
+  fs.writeFileSync(
+    queueFile,
+    JSON.stringify({
+      pending: [
+        {
+          key,
+          title: 'My Anime',
+          episode: 2,
+          createdAt: 1,
+          attemptCount: 0,
+          nextAttemptAt: 1,
+          lastError: null,
+        },
+      ],
+      deadLetter: [],
+    }),
+  );
+  const queue = createAnilistUpdateQueue(queueFile, loggerState.logger);
+  assert.equal(queue.nextReady()?.key, 'jellyfin://example.com/item/item::2');
+  assert.equal(fs.readFileSync(queueFile, 'utf8').includes('test-secret'), false);
+  queue.enqueue('unsafe', 'stream?api_key=test-secret', 3);
+  assert.equal(queue.getSnapshot().pending, 1);
+  queue.markSuccess(key);
+  assert.equal(queue.getSnapshot().pending, 0);
+});
+
+test('anilist retry queue discards empty normalized identities on load and enqueue', () => {
+  const queueFile = createTempQueueFile();
+  const loggerState = createLogger();
+  const invalidKeys = [
+    '',
+    '  ',
+    '::3',
+    'stream?api_key=secret::3',
+    'stream%3Fapi_key%3Dsecret::3',
+    'https://[invalid::3',
+  ];
+  const item = {
+    title: 'My Anime',
+    episode: 3,
+    createdAt: 1,
+    attemptCount: 0,
+    nextAttemptAt: 1,
+    lastError: null,
+  };
+  const validKey = 'https://example.com/Videos/item/stream?api_key=secret::3';
+  fs.writeFileSync(
+    queueFile,
+    JSON.stringify({
+      pending: [...invalidKeys, validKey].map((key) => ({ ...item, key })),
+      deadLetter: invalidKeys.map((key) => ({ ...item, key })),
+    }),
+  );
+  const queue = createAnilistUpdateQueue(queueFile, loggerState.logger);
+  assert.deepEqual(queue.getSnapshot(), { pending: 1, ready: 1, deadLetter: 0 });
+  const persisted = fs.readFileSync(queueFile, 'utf8');
+  assert.deepEqual(JSON.parse(persisted), {
+    pending: [{ ...item, key: 'jellyfin://example.com/item/item::3' }],
+    deadLetter: [],
+  });
+  for (const key of invalidKeys) {
+    queue.enqueue(key, 'My Anime', 3);
+    queue.markFailure(key, 'invalid');
+    queue.markSuccess(key);
+  }
+  assert.equal(fs.readFileSync(queueFile, 'utf8'), persisted);
+  queue.markFailure(validKey, 'retry', 10);
+  assert.equal(queue.nextReady(30_010)?.attemptCount, 1);
+  queue.markSuccess(validKey);
+  queue.enqueue(validKey, 'My Anime', 3);
+  assert.equal(queue.nextReady()?.key, 'jellyfin://example.com/item/item::3');
+});
+
 test('anilist update queue enqueues, snapshots, and dequeues success', () => {
   const queueFile = createTempQueueFile();
   const loggerState = createLogger();

src/core/services/anilist/anilist-update-queue.ts

@@ -1,5 +1,13 @@
 import * as fs from 'fs';
 import { ensureDirForFile } from '../../../shared/fs-utils';
+import { sanitizeMediaTitle, toMediaIdentityPath } from '../../../shared/media-identity';
+
+function normalizeAnilistRetryKey(key: string): string {
+  const parts = key.match(/^(.*)::(\d+)$/s);
+  const identity = toMediaIdentityPath(parts ? (parts[1] ?? '') : key);
+  if (!identity) return '';
+  return parts ? `${identity}::${parts[2]}` : identity;
+}
 
 const INITIAL_BACKOFF_MS = 30_000;
 const MAX_BACKOFF_MS = 6 * 60 * 60 * 1000;
@@ -105,6 +113,9 @@ export function createAnilistUpdateQueue(
             isValidPersistedMediaId(item.mediaId) &&
             (typeof item.lastError === 'string' || item.lastError === null),
         )
+        .filter((item) => sanitizeMediaTitle(item.title) !== null)
+        .map((item) => ({ ...item, key: normalizeAnilistRetryKey(item.key) }))
+        .filter((item) => item.key !== '')
         .slice(0, MAX_ITEMS);
       deadLetter = parsedDeadLetter
         .filter(
@@ -120,7 +131,11 @@ export function createAnilistUpdateQueue(
             isValidPersistedMediaId(item.mediaId) &&
             (typeof item.lastError === 'string' || item.lastError === null),
         )
+        .filter((item) => sanitizeMediaTitle(item.title) !== null)
+        .map((item) => ({ ...item, key: normalizeAnilistRetryKey(item.key) }))
+        .filter((item) => item.key !== '')
         .slice(0, MAX_ITEMS);
+      if (JSON.stringify({ pending, deadLetter }) !== JSON.stringify(parsed)) persist();
     } catch (error) {
       logger.error('Failed to load AniList retry queue.', error);
     }
@@ -136,6 +151,9 @@ export function createAnilistUpdateQueue(
       season: number | null = null,
       mediaId: number | null = null,
     ): void {
+      if (!sanitizeMediaTitle(title)) return;
+      key = normalizeAnilistRetryKey(key);
+      if (!key) return;
       const existing =
         pending.find((item) => item.key === key) || deadLetter.find((item) => item.key === key);
       if (existing) {
@@ -165,6 +183,8 @@ export function createAnilistUpdateQueue(
     },
 
     markSuccess(key: string): void {
+      key = normalizeAnilistRetryKey(key);
+      if (!key) return;
       const before = pending.length;
       pending = pending.filter((item) => item.key !== key);
       if (pending.length !== before) {
@@ -173,6 +193,8 @@ export function createAnilistUpdateQueue(
     },
 
     markFailure(key: string, reason: string, nowMs: number = Date.now()): void {
+      key = normalizeAnilistRetryKey(key);
+      if (!key) return;
       const item = pending.find((candidate) => candidate.key === key);
       if (!item) {
         return;

src/core/services/anilist/anilist-updater.test.ts

@@ -140,6 +140,62 @@ test('guessAnilistMediaInfo preserves useful guessit alternative title for ambig
   });
 });
 
+test('guessAnilistMediaInfo uses the display title for authenticated streams', async () => {
+  const targets: string[] = [];
+  const result = await guessAnilistMediaInfo(
+    'https://jellyfin.example/Videos/item/stream?static=true&api_key=test-secret',
+    'My Anime S02E03',
+    {
+      runGuessit: async (target) => {
+        targets.push(target);
+        throw new Error('use fallback parser');
+      },
+    },
+  );
+  assert.deepEqual(targets, ['My Anime S02E03']);
+  assert.deepEqual(result, {
+    title: 'My Anime',
+    season: 2,
+    episode: 3,
+    source: 'fallback',
+  });
+});
+
+test('guessAnilistMediaInfo preserves slashes in display titles', async () => {
+  const title = 'Fate/stay night S01E02';
+  for (const mediaPath of [null, title, 'https://example.com/stream?api_key=test-secret']) {
+    const targets: string[] = [];
+    await guessAnilistMediaInfo(mediaPath, title, {
+      runGuessit: async (target) => {
+        targets.push(target);
+        return JSON.stringify({ title: 'Fate/stay night', season: 1, episode: 2 });
+      },
+    });
+    assert.deepEqual(targets, [title]);
+  }
+});
+
+test('guessAnilistMediaInfo never parses stream URLs or their query-bearing filenames', async () => {
+  const unsafeInputs = [
+    'https://jellyfin.example/Videos/item/stream?static=true&api_key=test-secret',
+    'stream?static=true&api_key=test-secret&MediaSourceId=item',
+    'https://user:test-secret@example.com/video.mkv',
+  ];
+  for (const input of unsafeInputs) {
+    const targets: string[] = [];
+    const deps = {
+      runGuessit: async (target: string) => {
+        targets.push(target);
+        return JSON.stringify({ title: target });
+      },
+    };
+    assert.equal(await guessAnilistMediaInfo(input, null, deps), null);
+    assert.equal(await guessAnilistMediaInfo(null, input, deps), null);
+    assert.equal(await guessAnilistMediaInfo(input, input, deps), null);
+    assert.deepEqual(targets, []);
+  }
+});
+
 test('updateAnilistPostWatchProgress updates progress when behind', async () => {
   const originalFetch = globalThis.fetch;
   let call = 0;

src/core/services/anilist/anilist-updater.ts

@@ -2,6 +2,7 @@ import * as childProcess from 'child_process';
 import * as path from 'path';
 
 import { parseMediaInfo } from '../../../jimaku/utils';
+import { resolveMediaLookupTarget, sanitizeMediaTitle } from '../../../shared/media-identity';
 import type { AnilistRateLimiter } from './rate-limiter';
 import { resolveAnilistSeasonMedia } from './season-resolver';
 
@@ -231,8 +232,9 @@ export async function guessAnilistMediaInfo(
   mediaTitle: string | null,
   deps: GuessAnilistMediaInfoDeps = { runGuessit },
 ): Promise<AnilistMediaGuess | null> {
-  const target = mediaPath ?? mediaTitle;
-  const guessitTarget = mediaPath ? path.basename(mediaPath) : mediaTitle;
+  const target = resolveMediaLookupTarget(mediaPath, mediaTitle);
+  if (!target) return null;
+  const guessitTarget = target === sanitizeMediaTitle(mediaTitle) ? target : path.basename(target);
 
   if (guessitTarget && guessitTarget.trim().length > 0) {
     try {
@@ -260,8 +262,7 @@ export async function guessAnilistMediaInfo(
     }
   }
 
-  const fallbackTarget = mediaPath ?? mediaTitle;
-  const parsed = parseMediaInfo(fallbackTarget);
+  const parsed = parseMediaInfo(target);
   if (!parsed.title.trim()) {
     return null;
   }

src/core/services/anilist/cover-art-fetcher.test.ts

@@ -540,3 +540,200 @@ test('fetchIfMissing re-resolves an unresolved season once AniList publishes the
     cleanupDbPath(dbPath);
   }
 });
+
+for (const linkedToAnilist of [false, true]) {
+  test(`TMDB fallback preserves AniList identity when linked=${linkedToAnilist}`, async () => {
+    const dbPath = makeDbPath();
+    const db = new Database(dbPath);
+    ensureSchema(db);
+    const videoId = getOrCreateVideoRecord(db, 'local:/tmp/hanzawa-01.mkv', {
+      canonicalTitle: 'Hanzawa Naoki - 01.mkv',
+      sourcePath: '/tmp/hanzawa-01.mkv',
+      sourceUrl: null,
+      sourceType: SOURCE_TYPE_LOCAL,
+    });
+    const animeId = getOrCreateAnimeRecord(db, {
+      parsedTitle: 'Hanzawa Naoki',
+      canonicalTitle: 'Hanzawa Naoki',
+      anilistId: linkedToAnilist ? 42 : null,
+      titleRomaji: null,
+      titleEnglish: null,
+      titleNative: null,
+      metadataJson: null,
+    });
+    linkVideoToAnimeRecord(db, videoId, {
+      animeId,
+      parsedBasename: null,
+      parsedTitle: 'Hanzawa Naoki',
+      parsedSeason: null,
+      parsedEpisode: 1,
+      parserSource: 'fallback',
+      parserConfidence: 1,
+      parseMetadataJson: null,
+    });
+
+    const fetchCalls: string[] = [];
+    const originalFetch = globalThis.fetch;
+    globalThis.fetch = (async (input: RequestInfo | URL) => {
+      const url = String(input);
+      fetchCalls.push(url);
+      if (url.startsWith('https://graphql.anilist.co')) {
+        return createJsonResponse({ data: { Page: { media: [] } } });
+      }
+      assert.equal(url, 'https://image.tmdb.org/t/p/w500/hanzawa.jpg');
+      return new Response(new Uint8Array([5, 6, 7]), {
+        status: 200,
+        headers: { 'Content-Type': 'image/jpeg' },
+      });
+    }) as typeof fetch;
+
+    const resolvedTitles: string[] = [];
+    try {
+      const fetcher = createCoverArtFetcher(
+        { acquire: async () => {}, recordResponse: () => {} },
+        console,
+        {
+          runGuessit: async () => {
+            throw new Error('guessit unavailable');
+          },
+          liveAction: {
+            async resolveByTitle(title) {
+              resolvedTitles.push(title);
+              if (title !== 'Hanzawa Naoki') return null;
+              return {
+                tmdbId: 61222,
+                tmdbType: 'tv',
+                titleEnglish: 'Hanzawa Naoki',
+                titleNative: '半沢直樹',
+                description: 'A banker fights back.',
+                posterUrl: 'https://image.tmdb.org/t/p/w500/hanzawa.jpg',
+                episodesTotal: 10,
+                year: 2013,
+                originalLanguage: 'ja',
+                isAnimation: false,
+                allTitles: ['Hanzawa Naoki', '半沢直樹'],
+              };
+            },
+            async resolveById() {
+              return null;
+            },
+          },
+        },
+      );
+
+      const fetched = await fetcher.fetchIfMissing(db, videoId, 'Hanzawa Naoki - 01.mkv');
+      const stored = getCoverArt(db, videoId);
+      const anime = db
+        .prepare(
+          'SELECT media_kind AS mediaKind, tmdb_id AS tmdbId, description FROM imm_anime WHERE anime_id = ?',
+        )
+        .get(animeId) as { mediaKind: string; tmdbId: number | null; description: string | null };
+
+      if (linkedToAnilist) {
+        assert.equal(fetched, false);
+        assert.equal(stored?.coverBlob, null);
+        assert.equal(stored?.coverUrl, null);
+        assert.equal(anime.mediaKind, 'anime');
+        assert.equal(anime.tmdbId, null);
+        assert.deepEqual(resolvedTitles, []);
+        const requestCount = fetchCalls.length;
+        assert.equal(await fetcher.fetchIfMissing(db, videoId, 'Hanzawa Naoki - 01.mkv'), false);
+        assert.equal(fetchCalls.length, requestCount);
+        return;
+      }
+      assert.equal(fetched, true);
+      // The raw fallback-parser title is tried first, then the tag-stripped one.
+      assert.deepEqual(resolvedTitles, ['Hanzawa Naoki - 01', 'Hanzawa Naoki']);
+      assert.equal(stored?.anilistId, null);
+      assert.equal(stored?.coverUrl, 'https://image.tmdb.org/t/p/w500/hanzawa.jpg');
+      assert.equal(Buffer.from(stored?.coverBlob ?? []).toString('hex'), '050607');
+      assert.equal(anime.mediaKind, 'live_action');
+      assert.equal(anime.tmdbId, 61222);
+      assert.equal(anime.description, 'A banker fights back.');
+      assert.ok(fetchCalls.some((url) => url.startsWith('https://graphql.anilist.co')));
+    } finally {
+      globalThis.fetch = originalFetch;
+      db.close();
+      cleanupDbPath(dbPath);
+    }
+  });
+}
+
+test('fetchIfMissing skips AniList for an entry already linked to TMDB', async () => {
+  const dbPath = makeDbPath();
+  const db = new Database(dbPath);
+  ensureSchema(db);
+  const videoId = getOrCreateVideoRecord(db, 'local:/tmp/hanzawa-02.mkv', {
+    canonicalTitle: 'Hanzawa Naoki - 02.mkv',
+    sourcePath: '/tmp/hanzawa-02.mkv',
+    sourceUrl: null,
+    sourceType: SOURCE_TYPE_LOCAL,
+  });
+  const animeId = getOrCreateAnimeRecord(db, {
+    parsedTitle: 'Hanzawa Naoki',
+    canonicalTitle: 'Hanzawa Naoki',
+    anilistId: null,
+    titleRomaji: null,
+    titleEnglish: null,
+    titleNative: null,
+    metadataJson: null,
+  });
+  linkVideoToAnimeRecord(db, videoId, {
+    animeId,
+    parsedBasename: null,
+    parsedTitle: 'Hanzawa Naoki',
+    parsedSeason: null,
+    parsedEpisode: 2,
+    parserSource: 'fallback',
+    parserConfidence: 1,
+    parseMetadataJson: null,
+  });
+  db.prepare(
+    "UPDATE imm_anime SET media_kind = 'live_action', tmdb_id = 61222, tmdb_type = 'tv' WHERE anime_id = ?",
+  ).run(animeId);
+
+  const originalFetch = globalThis.fetch;
+  globalThis.fetch = (async (input: RequestInfo | URL) => {
+    assert.equal(String(input), 'https://image.tmdb.org/t/p/w500/hanzawa.jpg');
+    return new Response(new Uint8Array([1]), { status: 200 });
+  }) as typeof fetch;
+
+  const byIdCalls: Array<[string, number]> = [];
+  try {
+    const fetcher = createCoverArtFetcher(
+      { acquire: async () => {}, recordResponse: () => {} },
+      console,
+      {
+        liveAction: {
+          async resolveByTitle() {
+            throw new Error('title search must not run for a linked entry');
+          },
+          async resolveById(tmdbType, tmdbId) {
+            byIdCalls.push([tmdbType, tmdbId]);
+            return {
+              tmdbId,
+              tmdbType,
+              titleEnglish: 'Hanzawa Naoki',
+              titleNative: null,
+              description: null,
+              posterUrl: 'https://image.tmdb.org/t/p/w500/hanzawa.jpg',
+              episodesTotal: 10,
+              year: null,
+              originalLanguage: 'ja',
+              isAnimation: false,
+              allTitles: [],
+            };
+          },
+        },
+      },
+    );
+
+    assert.equal(await fetcher.fetchIfMissing(db, videoId, 'Hanzawa Naoki - 02.mkv'), true);
+    assert.deepEqual(byIdCalls, [['tv', 61222]]);
+    assert.equal(getCoverArt(db, videoId)?.coverBlob?.length, 1);
+  } finally {
+    globalThis.fetch = originalFetch;
+    db.close();
+    cleanupDbPath(dbPath);
+  }
+});

src/core/services/anilist/cover-art-fetcher.ts

@@ -17,6 +17,9 @@ import {
   type AnilistQueryExecutor,
   type AnilistSeasonResolution,
 } from './season-resolver';
+import { getVideoTmdbLink, linkAnimeToTmdbTitle } from '../immersion-tracker/live-action-link';
+import type { LiveActionMetadataResolver } from '../tmdb/live-action-resolver';
+import type { TmdbTitleDetails } from '../tmdb/tmdb-client';
 
 const ANILIST_GRAPHQL_URL = 'https://graphql.anilist.co';
 const NO_MATCH_RETRY_MS = 5 * 60 * 1000;
@@ -40,6 +43,8 @@ interface CoverArtCandidate {
 
 interface CoverArtFetcherOptions {
   runGuessit?: GuessAnilistMediaInfoDeps['runGuessit'];
+  /** Live-action fallback consulted when AniList has no match for a title. */
+  liveAction?: LiveActionMetadataResolver;
 }
 
 export function stripFilenameTags(raw: string): string {
@@ -153,6 +158,60 @@ export function createCoverArtFetcher(
     return true;
   };
 
+  const cacheNoMatch = (db: DatabaseSync, videoId: number): void => {
+    upsertCoverArt(db, videoId, {
+      anilistId: null,
+      coverUrl: null,
+      coverBlob: null,
+      titleRomaji: null,
+      titleEnglish: null,
+      episodesTotal: null,
+    });
+  };
+
+  // Links the video's library entry to the TMDB title and stores its poster.
+  const storeLiveActionArt = async (
+    db: DatabaseSync,
+    videoId: number,
+    details: TmdbTitleDetails,
+  ): Promise<boolean> => {
+    const row = db
+      .prepare(
+        `SELECT v.anime_id AS animeId, a.anilist_id AS anilistId
+                FROM imm_videos v LEFT JOIN imm_anime a ON a.anime_id = v.anime_id
+                WHERE v.video_id = ?`,
+      )
+      .get(videoId) as { animeId: number | null; anilistId: number | null } | undefined;
+    if (row?.anilistId != null) return false;
+    if (row?.animeId) {
+      const link = linkAnimeToTmdbTitle(db, row.animeId, details, { mode: 'auto' });
+      if (link.mergedAnimeIds.length > 0) {
+        logger.info(
+          'cover-art: folded library entries %s into %d (same TMDB title)',
+          link.mergedAnimeIds.join(','),
+          link.animeId,
+        );
+      }
+    }
+    const coverBlob = details.posterUrl ? await downloadImage(details.posterUrl) : null;
+    upsertCoverArt(db, videoId, {
+      anilistId: null,
+      coverUrl: details.posterUrl,
+      coverBlob,
+      titleRomaji: null,
+      titleEnglish: details.titleEnglish,
+      episodesTotal: details.episodesTotal,
+    });
+    logger.info(
+      'cover-art: linked videoId=%d to TMDB %s/%d "%s"',
+      videoId,
+      details.tmdbType,
+      details.tmdbId,
+      details.titleEnglish ?? details.titleNative ?? '',
+    );
+    return coverBlob !== null;
+  };
+
   const resolveCanonicalTitle = (
     db: DatabaseSync,
     videoId: number,
@@ -193,6 +252,16 @@ export function createCoverArtFetcher(
 
   return {
     async fetchIfMissing(db, videoId, canonicalTitle): Promise<boolean> {
+      const channel = db
+        .prepare(
+          `
+        SELECT 1 FROM imm_videos v
+        JOIN imm_anime a ON a.anime_id = v.anime_id
+        WHERE v.video_id = ? AND a.media_kind = 'youtube'
+      `,
+        )
+        .get(videoId);
+      if (channel) return false;
       const existing = getCoverArt(db, videoId);
       if (existing?.coverBlob) {
         return true;
@@ -226,18 +295,31 @@ export function createCoverArtFetcher(
         return false;
       }
 
+      // A live-action entry already knows its TMDB title; AniList has nothing
+      // to add and would only produce a spurious anime match.
+      const hasAnilistLink = Boolean(
+        db
+          .prepare(
+            `SELECT 1 FROM imm_videos v JOIN imm_anime a ON a.anime_id = v.anime_id
+         WHERE v.video_id = ? AND a.anilist_id IS NOT NULL`,
+          )
+          .get(videoId),
+      );
+      const tmdbLink = getVideoTmdbLink(db, videoId);
+      if (tmdbLink && !hasAnilistLink) {
+        const details = await options.liveAction?.resolveById(tmdbLink.tmdbType, tmdbLink.tmdbId);
+        if (details) {
+          return storeLiveActionArt(db, videoId, details);
+        }
+        cacheNoMatch(db, videoId);
+        return false;
+      }
+
       const effectiveTitle = resolveCanonicalTitle(db, videoId, canonicalTitle);
       const cleaned = stripFilenameTags(effectiveTitle);
       if (!cleaned) {
         logger.warn('cover-art: empty title after stripping tags for videoId=%d', videoId);
-        upsertCoverArt(db, videoId, {
-          anilistId: null,
-          coverUrl: null,
-          coverBlob: null,
-          titleRomaji: null,
-          titleEnglish: null,
-          episodesTotal: null,
-        });
+        cacheNoMatch(db, videoId);
         return false;
       }
 
@@ -295,15 +377,16 @@ export function createCoverArtFetcher(
 
       const selected = resolution?.media ?? null;
       if (!selected) {
-        logger.info('cover-art: no Anilist results for "%s", caching no-match', searchBase);
-        upsertCoverArt(db, videoId, {
-          anilistId: null,
-          coverUrl: null,
-          coverBlob: null,
-          titleRomaji: null,
-          titleEnglish: null,
-          episodesTotal: null,
-        });
+        if (options.liveAction && !hasAnilistLink) {
+          for (const searchTitle of searchTitles) {
+            const details = await options.liveAction.resolveByTitle(searchTitle);
+            if (details) {
+              return storeLiveActionArt(db, videoId, details);
+            }
+          }
+        }
+        logger.info('cover-art: no Anilist or TMDB results for "%s", caching no-match', searchBase);
+        cacheNoMatch(db, videoId);
         return false;
       }