feat(youtube): add mediaCache mode and safer stream media extraction (#130)

changes/youtube-media-generation-reliability.md

@@ -0,0 +1,7 @@
+type: fixed
+area: youtube
+
+- Improved YouTube card media generation by sending safer ffmpeg request options for resolved streams and skipping stale stream maps, including cached YouTube files.
+- Added `youtube.mediaCache.mode` with `direct` and `background` modes so YouTube card audio/image extraction can optionally use a background yt-dlp media cache when direct stream extraction is unreliable; background mode now announces cache download start/readiness through queued overlay/OSD notifications, creates text-only cards while the cache downloads, queues media updates for the mined note IDs, fills audio/image fields once the cached file is ready, and caps background downloads at `youtube.mediaCache.maxHeight` 720p by default while honoring height changes between downloads.
+- Cleaned stale YouTube background media cache files on startup and before new background downloads, and stopped in-flight background downloads when switching back to direct mode.
+- Hardened YouTube media cache downloads with IPv4/extractor retry flags and made failed background downloads notify users and clear queued media updates instead of leaving them pending silently.

config.example.jsonc

@@ -618,7 +618,11 @@
     "primarySubLanguages": [
       "ja",
       "jpn"
-    ] // Comma-separated primary subtitle language priority for managed subtitle auto-selection.
+    ], // Comma-separated primary subtitle language priority for managed subtitle auto-selection.
+    "mediaCache": {
+      "mode": "direct", // How YouTube card audio/images are extracted. Values: direct | background
+      "maxHeight": 720 // Maximum video height downloaded for the YouTube background media cache. Set to 0 for unlimited.
+    } // Media cache setting.
   }, // Defaults for managed subtitle language preferences and YouTube subtitle loading.
 
   // ==========================================

docs-site/configuration.md

@@ -1520,14 +1520,24 @@ Set defaults used by managed subtitle auto-selection and the `subminer` launcher
 ```json
 {
   "youtube": {
-    "primarySubLanguages": ["ja", "jpn"]
+    "primarySubLanguages": ["ja", "jpn"],
+    "mediaCache": {
+      "mode": "direct",
+      "maxHeight": 720
+    }
   }
 }
 ```
 
 | Option                 | Values                   | Description                                                                                      |
-| --------------------- | -------- | ------------------------------------------------------------------------------------------------ |
+| ---------------------- | ------------------------ | ------------------------------------------------------------------------------------------------ |
 | `primarySubLanguages`  | string[]                 | Primary subtitle language priority for managed subtitle auto-selection (default `["ja", "jpn"]`) |
+| `mediaCache.mode`      | `direct` \| `background` | YouTube card audio/image extraction mode (default `direct`)                                      |
+| `mediaCache.maxHeight` | number                   | Maximum background cache download height. Set `0` for unlimited (default `720`)                  |
+
+`mediaCache.mode: "direct"` extracts card media from the active YouTube stream URL. `mediaCache.mode: "background"` starts a separate yt-dlp media download after YouTube playback has loaded. Playback and subtitle loading do not wait for that download. Use background mode if direct card media generation hits YouTube `403` errors from expiring stream URLs.
+
+Background cache downloads are capped by `mediaCache.maxHeight`, which defaults to 720p; set it to `0` to let yt-dlp choose the best available height. Downloads use IPv4 and yt-dlp retry flags to reduce YouTube throttling failures. SubMiner announces when the background cache download starts and when the cache is ready, using the configured notification surface; overlay and OSD messages queue until the overlay or mpv is ready. If you mine cards before the cache is ready, SubMiner creates the text fields immediately, queues the audio/image work for those note IDs, shows a status notification, and fills the media fields once the cached file is ready. If the cache download fails, SubMiner shows a failure notification, shows queued-card failure notifications, and clears the pending updates.
 
 Current launcher behavior:
 

docs-site/public/config.example.jsonc

@@ -618,7 +618,11 @@
     "primarySubLanguages": [
       "ja",
       "jpn"
-    ] // Comma-separated primary subtitle language priority for managed subtitle auto-selection.
+    ], // Comma-separated primary subtitle language priority for managed subtitle auto-selection.
+    "mediaCache": {
+      "mode": "direct", // How YouTube card audio/images are extracted. Values: direct | background
+      "maxHeight": 720 // Maximum video height downloaded for the YouTube background media cache. Set to 0 for unlimited.
+    } // Media cache setting.
   }, // Defaults for managed subtitle language preferences and YouTube subtitle loading.
 
   // ==========================================

docs-site/youtube-integration.md

@@ -88,6 +88,12 @@ SubMiner handles several YouTube subtitle formats transparently:
 
 For auto-generated tracks, SubMiner prefers `srv3` > `srv2` > `srv1` > `vtt` (TimedText XML produces cleaner output). For manual tracks, `srt` > `vtt` is preferred.
 
+## Card Media Cache
+
+By default, YouTube card audio and screenshots are extracted directly from mpv's active stream URLs. If generated card media fails with YouTube `403` errors, set `youtube.mediaCache.mode` to `"background"`. Background mode starts a separate `yt-dlp` media download after playback loads, creates text fields immediately, queues audio/image work for mined notes, and fills those fields once the local cache file is ready.
+
+Background cache downloads use IPv4 and retry flags to reduce YouTube throttling failures. If the background download still fails, SubMiner shows a cache failure notification, shows queued-card failure notifications, and clears those pending updates so cards are not left waiting silently.
+
 ## Configuration Reference
 
 ### Primary Subtitle Languages
@@ -131,6 +137,7 @@ Precedence: CLI flag > environment variable > `config.jsonc` > built-in default.
 - **No subtitles found**: The video may not have Japanese subtitles. Open the picker with `Ctrl+Alt+C` to see all available tracks.
 - **yt-dlp not found**: Install `yt-dlp` and ensure it is on `PATH`, or set `SUBMINER_YTDLP_BIN` to the binary path.
 - **Probe timeout**: `yt-dlp` has a 15-second timeout per operation. Slow connections or rate-limited IPs may hit this. Retry or update `yt-dlp`.
+- **Card media `403` errors**: Switch `youtube.mediaCache.mode` from `"direct"` to `"background"` so card media is generated from a local `yt-dlp` cache instead of ffmpeg reading an expiring YouTube stream URL.
 - **Auto-caption quality**: YouTube auto-generated captions vary in quality. Manual subtitles (when available) are always preferred.
 - **`ytsearch:` targets**: `subminer ytsearch:"keyword"` plays the first search result. Subtitle availability depends on the matched video.
 - **Secondary subtitle fails**: Secondary track failures never block playback. The primary subtitle loads independently.

release/release-notes.md

@@ -1,68 +0,0 @@
-## Highlights
-### Changed
-
-- **Subtitle Delay Keybindings:** Overlay subtitle delay controls now match mpv's native bindings.
-  - `z`, `Z`, and `x` adjust subtitle delay; `Ctrl+Shift+Left/Right` step to the adjacent subtitle and show the current delay on the OSD.
-  - Removes the old SubMiner-specific adjacent-cue delay action in favor of mpv's built-in `sub-step`.
-
-- **Update Notifications:** New installs now default to overlay-only update notifications instead of also sending a system notification.
-
-### Fixed
-
-- **Anki — Highlight Word:** The mined word is now correctly bolded in Kiku sentence and sentence-furigana fields even when the source Yomitan sentence did not already contain bold markup.
-
-- **Anki — Lapis/Kiku Word Cards:** Word cards enriched through SubMiner now correctly include the word-and-sentence marker, restoring sentence context on the card front.
-
-- **Anki — Windows:** Fixed two issues that surfaced after background app launches on Windows.
-  - Audio clip and image generation now works correctly by recreating missing FFmpeg temp directories before export.
-  - Known-word cache refreshes no longer fail when no deck is configured.
-
-- **Desktop Notifications:** Restored the SubMiner app icon on system notifications that do not supply their own image.
-
-- **Dictionary — Windows:** The character dictionary auto-sync on `SubMiner mpv` shortcut launches can now fall back to mpv's current video path when app media state is not yet ready.
-
-- **Overlay — macOS Yomitan Popup:** Fixed focus and dismiss behavior for the Yomitan popup on macOS.
-  - Popup focus is correctly restored after mining a card or reloading the popup.
-  - Clicking on transparent overlay space now properly closes the popup and passes the click through to mpv, with no hide/reappear cycle.
-
-- **Overlay — Linux Startup:** Fixed several edge cases that could leave the overlay unresponsive or drop subtitles at startup when auto-pause was active.
-  - The overlay stays interactive during the initial render measurement gap.
-  - Subtitles paint as plain text immediately on cache misses, before tokenization finishes.
-  - Temporarily empty subtitle state is now re-parsed correctly before warm readiness resumes playback.
-
-- **Overlay — Playlist Advance:** The visible overlay now stays interactive when mpv advances to the next playlist item, including when the next episode loads after the warm transition delay.
-
-- **Overlay — Windows:** Fixed shaky subtitle-bar hover and click behavior when a video connects to an already-running background SubMiner instance.
-
-- **Stats — AniList Search:** Manual AniList linking from the stats anime page now searches only the anime title, dropping any generated "Season N" suffix that was causing failed lookups.
-
-- **Updates — Linux:** Improved Linux update reliability for managed support assets.
-  - Updates now correctly install and refresh both the launcher runtime plugin copy and the rofi theme alongside AppImage and launcher updates.
-  - Support-asset refreshes no longer touch unrelated SubMiner data directories, and plugin copies are staged safely before replacing the live runtime plugin.
-  - Fresh installs now auto-install the managed runtime plugin and rofi theme from the bundled app on first launcher playback if either asset is missing.
-
-### Docs
-
-- **Linux Updates:** Documented how Linux update flows manage the launcher runtime plugin and rofi theme, and that the first launcher playback auto-installs any missing managed support assets from the bundled app.
-
-## What's Changed
-
-- Replace subtitle delay actions with native mpv keybindings by @ksyasuda in #120
-- fix(stats): strip Season N suffix from AniList title searches by @ksyasuda in #121
-- fix(overlay): preserve visible state across playlist item transitions by @ksyasuda in #124
-- fix(overlay): restore macOS Yomitan popup focus without breaking click-away by @ksyasuda in #125
-- fix(linux): auto-install managed plugin copy; include in asset updates by @ksyasuda in #127
-- Fix Windows Anki startup and overlay regressions by @ksyasuda in #128
-
-## Installation
-
-See the README and docs/installation guide for full setup steps.
-
-## Assets
-
-- Linux: `SubMiner.AppImage`
-- macOS: `SubMiner-*.dmg` and `SubMiner-*.zip`
-- Windows: `SubMiner-*.exe` and `SubMiner-*-win.zip`
-- Optional extras: `subminer-assets.tar.gz` and the `subminer` launcher
-
-Note: the `subminer` wrapper script uses Bun (`#!/usr/bin/env bun`), so `bun` must be installed and on `PATH`.

src/anki-integration.test.ts

@@ -5,6 +5,7 @@ import * as os from 'os';
 import * as path from 'path';
 import { AnkiIntegration } from './anki-integration';
 import { FieldGroupingMergeCollaborator } from './anki-integration/field-grouping-merge';
+import type { MediaInput } from './media-input';
 import { AnkiConnectConfig } from './types';
 
 type TestOverlayNotificationPayload = {
@@ -24,6 +25,13 @@ interface IntegrationTestContext {
   stateDir: string;
 }
 
+function describeMediaInputForTest(input: MediaInput): string {
+  if (typeof input === 'string') {
+    return input;
+  }
+  return `${input.path}:${input.source ?? 'raw'}`;
+}
+
 function createIntegrationTestContext(
   options: {
     highlightEnabled?: boolean;
@@ -527,6 +535,354 @@ test('AnkiIntegration marks partial update notifications as failures in OSD mode
   assert.deepEqual(osdMessages, ['x Updated card: taberu (image failed)']);
 });
 
+test('AnkiIntegration applies ready YouTube cache media to every queued note id', async () => {
+  const osdMessages: string[] = [];
+  const updatedNotes: Array<{ noteId: number; fields: Record<string, string> }> = [];
+  const storedMedia: string[] = [];
+  const mediaInputs: string[] = [];
+
+  const integration = new AnkiIntegration(
+    {
+      fields: {
+        image: 'Picture',
+      },
+      media: {
+        imageFormat: 'jpg',
+      },
+      behavior: {
+        notificationType: 'osd',
+      },
+    },
+    {} as never,
+    {} as never,
+    (text) => {
+      osdMessages.push(text);
+    },
+  );
+
+  const internals = integration as unknown as {
+    client: {
+      notesInfo: (noteIds: number[]) => Promise<unknown[]>;
+      updateNoteFields: (noteId: number, fields: Record<string, string>) => Promise<void>;
+      storeMediaFile: (filename: string, data: Buffer) => Promise<void>;
+    };
+    mediaGenerator: {
+      generateAudio: (
+        path: MediaInput,
+        startTime: number,
+        endTime: number,
+        audioPadding?: number,
+        audioStreamIndex?: number,
+      ) => Promise<Buffer>;
+      generateScreenshot: (path: MediaInput) => Promise<Buffer>;
+    };
+    queuePendingYoutubeMediaUpdate: (job: {
+      sourceUrl: string;
+      noteId: number;
+      startTime: number;
+      endTime: number;
+      label: string | number;
+      audioStreamIndex?: number;
+      audioFieldName?: string;
+      imageFieldName?: string;
+      generateAudio: boolean;
+      generateImage: boolean;
+    }) => void;
+  };
+  internals.client = {
+    notesInfo: async (noteIds) =>
+      noteIds.map((noteId) => ({
+        noteId,
+        fields: {
+          SentenceAudio: { value: '' },
+          Picture: { value: '' },
+        },
+      })),
+    updateNoteFields: async (noteId, fields) => {
+      updatedNotes.push({ noteId, fields });
+    },
+    storeMediaFile: async (filename) => {
+      storedMedia.push(filename);
+    },
+  };
+  internals.mediaGenerator = {
+    generateAudio: async (mediaPath, _startTime, _endTime, _audioPadding, audioStreamIndex) => {
+      mediaInputs.push(
+        `audio:${describeMediaInputForTest(mediaPath)}:${audioStreamIndex ?? 'auto'}`,
+      );
+      return Buffer.from('audio');
+    },
+    generateScreenshot: async (mediaPath) => {
+      mediaInputs.push(`image:${describeMediaInputForTest(mediaPath)}`);
+      return Buffer.from('image');
+    },
+  };
+  internals.queuePendingYoutubeMediaUpdate({
+    sourceUrl: 'https://www.youtube.com/watch?v=abc123',
+    noteId: 101,
+    startTime: 10,
+    endTime: 12,
+    label: 'first',
+    audioStreamIndex: 22,
+    audioFieldName: 'SentenceAudio',
+    imageFieldName: 'Picture',
+    generateAudio: true,
+    generateImage: true,
+  });
+  internals.queuePendingYoutubeMediaUpdate({
+    sourceUrl: 'https://youtu.be/abc123',
+    noteId: 202,
+    startTime: 20,
+    endTime: 22,
+    label: 'second',
+    audioStreamIndex: 23,
+    audioFieldName: 'SentenceAudio',
+    imageFieldName: 'Picture',
+    generateAudio: true,
+    generateImage: true,
+  });
+
+  await integration.handleYoutubeMediaCacheReady('https://youtu.be/abc123', '/tmp/media.mkv');
+
+  assert.deepEqual(mediaInputs, [
+    'audio:/tmp/media.mkv:youtube-cache:auto',
+    'image:/tmp/media.mkv:youtube-cache',
+    'audio:/tmp/media.mkv:youtube-cache:auto',
+    'image:/tmp/media.mkv:youtube-cache',
+  ]);
+  assert.deepEqual(
+    updatedNotes.map((update) => update.noteId),
+    [101, 202],
+  );
+  const firstUpdate = updatedNotes[0];
+  const secondUpdate = updatedNotes[1];
+  assert.ok(firstUpdate);
+  assert.ok(secondUpdate);
+  assert.match(firstUpdate.fields.SentenceAudio ?? '', /^\[sound:audio_/);
+  assert.match(firstUpdate.fields.Picture ?? '', /^<img src="image_/);
+  assert.match(secondUpdate.fields.SentenceAudio ?? '', /^\[sound:audio_/);
+  assert.match(secondUpdate.fields.Picture ?? '', /^<img src="image_/);
+  assert.equal(storedMedia.length, 4);
+  assert.equal(
+    osdMessages.some((message) =>
+      message.includes('YouTube media cache ready. Adding media to 2 queued cards.'),
+    ),
+    true,
+  );
+});
+
+test('AnkiIntegration reports partial queued YouTube media updates separately from failures', async () => {
+  const osdMessages: string[] = [];
+  const updatedNotes: Array<{ noteId: number; fields: Record<string, string> }> = [];
+  const notifications: Array<{ noteId: number; label: string | number; suffix?: string }> = [];
+
+  const integration = new AnkiIntegration(
+    {
+      fields: {
+        image: 'Picture',
+      },
+      media: {
+        imageFormat: 'jpg',
+      },
+      behavior: {
+        notificationType: 'osd',
+      },
+    },
+    {} as never,
+    {} as never,
+    (text) => {
+      osdMessages.push(text);
+    },
+  );
+
+  const internals = integration as unknown as {
+    client: {
+      notesInfo: (noteIds: number[]) => Promise<unknown[]>;
+      updateNoteFields: (noteId: number, fields: Record<string, string>) => Promise<void>;
+      storeMediaFile: () => Promise<void>;
+    };
+    mediaGenerator: {
+      generateAudio: () => Promise<Buffer>;
+      generateScreenshot: () => Promise<Buffer>;
+    };
+    queuePendingYoutubeMediaUpdate: (job: {
+      sourceUrl: string;
+      noteId: number;
+      startTime: number;
+      endTime: number;
+      label: string | number;
+      audioFieldName?: string;
+      imageFieldName?: string;
+      generateAudio: boolean;
+      generateImage: boolean;
+    }) => void;
+    showNotification: (noteId: number, label: string | number, suffix?: string) => Promise<void>;
+  };
+  internals.client = {
+    notesInfo: async (noteIds) =>
+      noteIds.map((noteId) => ({
+        noteId,
+        fields: {
+          SentenceAudio: { value: '' },
+          Picture: { value: '' },
+        },
+      })),
+    updateNoteFields: async (noteId, fields) => {
+      updatedNotes.push({ noteId, fields });
+    },
+    storeMediaFile: async () => undefined,
+  };
+  internals.mediaGenerator = {
+    generateAudio: async () => {
+      throw new Error('audio stream not found');
+    },
+    generateScreenshot: async () => Buffer.from('image'),
+  };
+  internals.showNotification = async (noteId, label, suffix) => {
+    notifications.push({ noteId, label, suffix });
+  };
+
+  internals.queuePendingYoutubeMediaUpdate({
+    sourceUrl: 'https://www.youtube.com/watch?v=partial',
+    noteId: 303,
+    startTime: 10,
+    endTime: 12,
+    label: 'partial',
+    audioFieldName: 'SentenceAudio',
+    imageFieldName: 'Picture',
+    generateAudio: true,
+    generateImage: true,
+  });
+
+  await integration.handleYoutubeMediaCacheReady('https://youtu.be/partial', '/tmp/media.mkv');
+
+  assert.equal(updatedNotes.length, 1);
+  assert.match(updatedNotes[0]?.fields.Picture ?? '', /^<img src="image_/);
+  assert.equal(updatedNotes[0]?.fields.SentenceAudio, undefined);
+  assert.deepEqual(notifications, [{ noteId: 303, label: 'partial', suffix: 'audio failed' }]);
+  assert.equal(
+    osdMessages.some((message) =>
+      message.includes('Queued YouTube media finished with 0 updated, 1 partial, and 0 failed.'),
+    ),
+    true,
+  );
+});
+
+test('AnkiIntegration does not use mpv stream indexes for ready cached YouTube audio', async () => {
+  const audioCalls: Array<{ path: string; audioStreamIndex?: number }> = [];
+
+  const integration = new AnkiIntegration(
+    {
+      media: {
+        audioPadding: 0,
+      },
+    },
+    {} as never,
+    {
+      currentVideoPath: 'https://www.youtube.com/watch?v=abc123',
+      currentAudioStreamIndex: 0,
+      currentSubStart: 10,
+      currentSubEnd: 12,
+      currentTimePos: 11,
+    } as never,
+    () => undefined,
+    undefined,
+    undefined,
+    undefined,
+    {},
+    undefined,
+    undefined,
+    async () => '/tmp/subminer-youtube-media-cache/media.mkv',
+    () => true,
+  );
+
+  const internals = integration as unknown as {
+    mediaGenerator: {
+      generateAudio: (
+        path: { path: string },
+        startTime: number,
+        endTime: number,
+        audioPadding?: number,
+        audioStreamIndex?: number,
+      ) => Promise<Buffer>;
+    };
+    generateAudio: () => Promise<Buffer | null>;
+  };
+  internals.mediaGenerator = {
+    generateAudio: async (path, _startTime, _endTime, _audioPadding, audioStreamIndex) => {
+      audioCalls.push({ path: path.path, audioStreamIndex });
+      return Buffer.from('audio');
+    },
+  };
+
+  await internals.generateAudio();
+
+  assert.deepEqual(audioCalls, [
+    {
+      path: '/tmp/subminer-youtube-media-cache/media.mkv',
+      audioStreamIndex: undefined,
+    },
+  ]);
+});
+
+test('AnkiIntegration announces ready YouTube cache when no queued notes exist', async () => {
+  const overlayNotifications: TestOverlayNotificationPayload[] = [];
+
+  const integration = new AnkiIntegration(
+    {
+      behavior: {
+        notificationType: 'overlay',
+      },
+    },
+    {} as never,
+    {} as never,
+    undefined,
+    undefined,
+    undefined,
+    undefined,
+    {},
+    undefined,
+    (payload) => {
+      overlayNotifications.push(payload as TestOverlayNotificationPayload);
+    },
+  );
+
+  await integration.handleYoutubeMediaCacheReady('https://youtu.be/abc123', '/tmp/media.mkv');
+
+  assert.equal(overlayNotifications.length, 1);
+  assert.equal(overlayNotifications[0]?.title, 'SubMiner');
+  assert.equal(overlayNotifications[0]?.body, 'YouTube media cache ready.');
+});
+
+test('AnkiIntegration can let caller own no-queued YouTube cache ready notification', async () => {
+  const overlayNotifications: TestOverlayNotificationPayload[] = [];
+
+  const integration = new AnkiIntegration(
+    {
+      behavior: {
+        notificationType: 'overlay',
+      },
+    },
+    {} as never,
+    {} as never,
+    undefined,
+    undefined,
+    undefined,
+    undefined,
+    {},
+    undefined,
+    (payload) => {
+      overlayNotifications.push(payload as TestOverlayNotificationPayload);
+    },
+  );
+
+  await integration.handleYoutubeMediaCacheReady('https://youtu.be/abc123', '/tmp/media.mkv', {
+    notifyNoQueued: false,
+  });
+
+  assert.deepEqual(overlayNotifications, []);
+});
+
 test('AnkiIntegration embeds generated notification image on overlay mined-card notifications', async () => {
   const desktopNotifications: Array<{ title: string; body?: string; icon?: string }> = [];
   const overlayNotifications: TestOverlayNotificationPayload[] = [];

src/anki-integration.ts

@@ -61,7 +61,18 @@ import { NoteUpdateWorkflow } from './anki-integration/note-update-workflow';
 import { FieldGroupingWorkflow } from './anki-integration/field-grouping-workflow';
 import { resolveAnimatedImageLeadInSeconds } from './anki-integration/animated-image-sync';
 import { AnkiIntegrationRuntime, normalizeAnkiIntegrationConfig } from './anki-integration/runtime';
-import { resolveMediaGenerationInputPath } from './anki-integration/media-source';
+import {
+  resolveAudioStreamIndexForMediaGeneration,
+  resolveMediaGenerationInput,
+  resolveMediaGenerationInputPath,
+  type MediaGenerationInputResolverOptions,
+} from './anki-integration/media-source';
+import type { PendingYoutubeMediaUpdate } from './anki-integration/pending-youtube-media';
+import { PendingYoutubeMediaQueue } from './anki-integration/pending-youtube-media-queue';
+import type {
+  PendingYoutubeMediaQueueFailedOptions,
+  PendingYoutubeMediaQueueReadyOptions,
+} from './anki-integration/pending-youtube-media-queue';
 
 const log = createLogger('anki').child('integration');
 
@@ -225,6 +236,10 @@ export class AnkiIntegration {
   private consumeSubtitleMiningContextCallback: (() => SubtitleMiningContext | null) | null = null;
   private noteIdRedirects = new Map<number, number>();
   private trackedDuplicateNoteIds = new Map<number, number[]>();
+  private getCachedMediaPath: MediaGenerationInputResolverOptions['getCachedMediaPath'] | null =
+    null;
+  private shouldRequireRemoteMediaCache: (() => boolean) | null = null;
+  private pendingYoutubeMediaQueue: PendingYoutubeMediaQueue;
 
   constructor(
     config: AnkiConnectConfig,
@@ -240,6 +255,8 @@ export class AnkiIntegration {
     aiConfig: AiConfig = {},
     recordCardsMined?: (count: number, noteIds?: number[]) => void,
     overlayNotificationCallback?: (payload: OverlayNotificationPayload) => void,
+    getCachedMediaPath?: MediaGenerationInputResolverOptions['getCachedMediaPath'],
+    shouldRequireRemoteMediaCache?: () => boolean,
   ) {
     this.config = normalizeAnkiIntegrationConfig(config);
     this.aiConfig = { ...aiConfig };
@@ -252,6 +269,9 @@ export class AnkiIntegration {
     this.overlayNotificationCallback = overlayNotificationCallback || null;
     this.fieldGroupingCallback = fieldGroupingCallback || null;
     this.recordCardsMinedCallback = recordCardsMined ?? null;
+    this.getCachedMediaPath = getCachedMediaPath ?? null;
+    this.shouldRequireRemoteMediaCache = shouldRequireRemoteMediaCache ?? null;
+    this.pendingYoutubeMediaQueue = this.createPendingYoutubeMediaQueue();
     this.knownWordCache = this.createKnownWordCache(knownWordCacheStatePath);
     this.pollingRunner = this.createPollingRunner();
     this.cardCreationService = this.createCardCreationService();
@@ -295,6 +315,81 @@ export class AnkiIntegration {
     }
   }
 
+  private getMediaResolverOptions(): MediaGenerationInputResolverOptions {
+    const options: MediaGenerationInputResolverOptions = {
+      logDebug: (message) => log.debug(message),
+    };
+    if (this.getCachedMediaPath) {
+      options.getCachedMediaPath = this.getCachedMediaPath;
+    }
+    if (this.shouldRequireRemoteMediaCache?.()) {
+      options.remoteCacheMode = 'required';
+    }
+    return options;
+  }
+
+  private createPendingYoutubeMediaQueue(): PendingYoutubeMediaQueue {
+    return new PendingYoutubeMediaQueue({
+      client: {
+        notesInfo: async (noteIds) => (await this.client.notesInfo(noteIds)) as unknown,
+        updateNoteFields: (noteId, fields) => this.client.updateNoteFields(noteId, fields),
+        storeMediaFile: (filename, data) => this.client.storeMediaFile(filename, data),
+      },
+      mediaGenerator: {
+        generateAudio: (videoPath, startTime, endTime, audioPadding, audioStreamIndex) =>
+          this.mediaGenerator.generateAudio(
+            videoPath,
+            startTime,
+            endTime,
+            audioPadding,
+            audioStreamIndex,
+          ),
+        generateScreenshot: (videoPath, timestamp, options) =>
+          this.mediaGenerator.generateScreenshot(videoPath, timestamp, options),
+        generateAnimatedImage: (videoPath, startTime, endTime, audioPadding, options) =>
+          this.mediaGenerator.generateAnimatedImage(
+            videoPath,
+            startTime,
+            endTime,
+            audioPadding,
+            options,
+          ),
+      },
+      getConfig: () => this.config,
+      getCurrentVideoPath: () => this.mpvClient.currentVideoPath,
+      getCachedMediaPath: this.getCachedMediaPath,
+      shouldRequireRemoteMediaCache: () => this.shouldRequireRemoteMediaCache?.() === true,
+      getSubtitleMediaRange: (context) => this.getSubtitleMediaRange(context),
+      getResolvedSentenceAudioFieldName: (noteInfo) =>
+        this.getResolvedSentenceAudioFieldName(noteInfo),
+      resolveConfiguredFieldName: (noteInfo, ...preferredNames) =>
+        this.resolveConfiguredFieldName(noteInfo, ...preferredNames),
+      mergeFieldValue: (existing, newValue, overwrite) =>
+        this.mergeFieldValue(existing, newValue, overwrite),
+      getAnimatedImageLeadInSeconds: (noteInfo) => this.getAnimatedImageLeadInSeconds(noteInfo),
+      generateAudioFilename: () => this.generateAudioFilename(),
+      generateImageFilename: () => this.generateImageFilename(),
+      formatMiscInfoPatternForMediaPath: (
+        fallbackFilename,
+        startTimeSeconds,
+        mediaPath,
+        mediaTitle,
+      ) =>
+        this.formatMiscInfoPatternForMediaPath(
+          fallbackFilename,
+          startTimeSeconds,
+          mediaPath,
+          mediaTitle,
+        ),
+      showStatusNotification: (message) => this.showStatusNotification(message),
+      showNotification: (noteId, label, errorSuffix) =>
+        this.showNotification(noteId, label, errorSuffix),
+      logInfo: (...args) => log.info(args[0] as string, ...args.slice(1)),
+      logWarn: (...args) => log.warn(args[0] as string, ...args.slice(1)),
+      logError: (...args) => log.error(args[0] as string, ...args.slice(1)),
+    });
+  }
+
   private createKnownWordCache(knownWordCacheStatePath?: string): KnownWordCacheManager {
     return new KnownWordCacheManager({
       client: {
@@ -377,6 +472,9 @@ export class AnkiIntegration {
       getAiConfig: () => this.aiConfig,
       getTimingTracker: () => this.timingTracker,
       getMpvClient: () => this.mpvClient,
+      ...(this.getCachedMediaPath ? { getCachedMediaPath: this.getCachedMediaPath } : {}),
+      shouldRequireRemoteMediaCache: () => this.shouldRequireRemoteMediaCache?.() === true,
+      queuePendingYoutubeMediaUpdate: (job) => this.queuePendingYoutubeMediaUpdate(job),
       getDeck: () => this.config.deck,
       client: {
         addNote: (deck, modelName, fields, tags) =>
@@ -540,6 +638,7 @@ export class AnkiIntegration {
       formatMiscInfoPattern: (fallbackFilename, startTimeSeconds) =>
         this.formatMiscInfoPattern(fallbackFilename, startTimeSeconds),
       consumeSubtitleMiningContext: () => this.consumeSubtitleMiningContext(),
+      queuePendingYoutubeMediaUpdate: (job) => this.queuePendingYoutubeMediaUpdateForNote(job),
       addConfiguredTagsToNote: (noteId) => this.addConfiguredTagsToNote(noteId),
       showNotification: (noteId, label) => this.showNotification(noteId, label),
       showOsdNotification: (message) => this.showStatusNotification(message),
@@ -833,13 +932,45 @@ export class AnkiIntegration {
     };
   }
 
+  private queuePendingYoutubeMediaUpdate(job: PendingYoutubeMediaUpdate): void {
+    this.pendingYoutubeMediaQueue.enqueue(job);
+  }
+
+  private async queuePendingYoutubeMediaUpdateForNote(job: {
+    noteId: number;
+    noteInfo: NoteInfo;
+    context?: SubtitleMiningContext;
+    label: string | number;
+  }): Promise<boolean> {
+    return this.pendingYoutubeMediaQueue.queueFromNote(job);
+  }
+
+  async handleYoutubeMediaCacheReady(
+    sourceUrl: string,
+    cachedPath: string,
+    options?: PendingYoutubeMediaQueueReadyOptions,
+  ): Promise<void> {
+    await this.pendingYoutubeMediaQueue.handleReady(sourceUrl, cachedPath, options);
+  }
+
+  async handleYoutubeMediaCacheFailed(
+    sourceUrl: string,
+    options?: PendingYoutubeMediaQueueFailedOptions,
+  ): Promise<void> {
+    await this.pendingYoutubeMediaQueue.handleFailed(sourceUrl, options);
+  }
+
   private async generateAudio(context?: SubtitleMiningContext): Promise<Buffer | null> {
     const mpvClient = this.mpvClient;
     if (!mpvClient || !mpvClient.currentVideoPath) {
       return null;
     }
 
-    const videoPath = await resolveMediaGenerationInputPath(mpvClient, 'audio');
+    const videoPath = await resolveMediaGenerationInput(
+      mpvClient,
+      'audio',
+      this.getMediaResolverOptions(),
+    );
     if (!videoPath) {
       return null;
     }
@@ -850,7 +981,7 @@ export class AnkiIntegration {
       startTime,
       endTime,
       this.config.media?.audioPadding,
-      this.mpvClient.currentAudioStreamIndex,
+      resolveAudioStreamIndexForMediaGeneration(videoPath, this.mpvClient.currentAudioStreamIndex),
     );
   }
 
@@ -862,7 +993,11 @@ export class AnkiIntegration {
       return null;
     }
 
-    const videoPath = await resolveMediaGenerationInputPath(this.mpvClient, 'video');
+    const videoPath = await resolveMediaGenerationInput(
+      this.mpvClient,
+      'video',
+      this.getMediaResolverOptions(),
+    );
     if (!videoPath) {
       return null;
     }
@@ -896,17 +1031,30 @@ export class AnkiIntegration {
   }
 
   private formatMiscInfoPattern(fallbackFilename: string, startTimeSeconds?: number): string {
+    return this.formatMiscInfoPatternForMediaPath(
+      fallbackFilename,
+      startTimeSeconds,
+      this.mpvClient.currentVideoPath || '',
+      this.mpvClient.currentMediaTitle ?? undefined,
+    );
+  }
+
+  private formatMiscInfoPatternForMediaPath(
+    fallbackFilename: string,
+    startTimeSeconds: number | undefined,
+    mediaPath: string,
+    mediaTitle?: string,
+  ): string {
     if (!this.config.metadata?.pattern) {
       return '';
     }
 
-    const currentVideoPath = this.mpvClient.currentVideoPath || '';
-    const videoFilename = extractFilenameFromMediaPath(currentVideoPath);
-    const mediaTitle = trimToNonEmptyString(this.mpvClient.currentMediaTitle);
+    const videoFilename = extractFilenameFromMediaPath(mediaPath);
+    const resolvedMediaTitle = trimToNonEmptyString(mediaTitle);
     const filenameWithExt =
-      (shouldPreferMediaTitleForMiscInfo(currentVideoPath, videoFilename)
-        ? mediaTitle || videoFilename
-        : videoFilename || mediaTitle) || fallbackFilename;
+      (shouldPreferMediaTitleForMiscInfo(mediaPath, videoFilename)
+        ? resolvedMediaTitle || videoFilename
+        : videoFilename || resolvedMediaTitle) || fallbackFilename;
     const filenameWithoutExt = filenameWithExt.replace(/\.[^.]+$/, '');
 
     const currentTimePos =
@@ -1457,7 +1605,7 @@ export class AnkiIntegration {
       miscInfoValue?: string;
     } = {};
 
-    if (this.config.media?.generateAudio && this.mpvClient?.currentVideoPath) {
+    if (this.config.media?.generateAudio !== false && this.mpvClient?.currentVideoPath) {
       try {
         const audioFilename = this.generateAudioFilename();
         const audioBuffer = await this.generateAudio();
@@ -1477,7 +1625,7 @@ export class AnkiIntegration {
       }
     }
 
-    if (this.config.media?.generateImage && this.mpvClient?.currentVideoPath) {
+    if (this.config.media?.generateImage !== false && this.mpvClient?.currentVideoPath) {
       try {
         const animatedLeadInSeconds = noteInfo
           ? await this.getAnimatedImageLeadInSeconds(noteInfo)

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

@@ -2,6 +2,7 @@ import assert from 'node:assert/strict';
 import test from 'node:test';
 
 import { CardCreationService } from './card-creation';
+import type { MediaInput } from '../media-generator';
 import type { AnkiConnectConfig } from '../types/anki';
 
 type CardCreationDeps = ConstructorParameters<typeof CardCreationService>[0];
@@ -266,6 +267,8 @@ test('manual clipboard subtitle update skips audio when sentence audio field is
 test('manual clipboard subtitle update uses resolved mpv stream URLs for remote media', async () => {
   const audioPaths: string[] = [];
   const imagePaths: string[] = [];
+  const recordMediaPath = (mediaInput: MediaInput): string =>
+    typeof mediaInput === 'string' ? mediaInput : mediaInput.path;
   const edlSource = [
     'edl://!new_stream;!no_clip;!no_chapters;%70%https://audio.example/videoplayback?mime=audio%2Fwebm',
     '!new_stream;!no_clip;!no_chapters;%69%https://video.example/videoplayback?mime=video%2Fmp4',
@@ -338,11 +341,11 @@ test('manual clipboard subtitle update uses resolved mpv stream URLs for remote
     },
     mediaGenerator: {
       generateAudio: async (path) => {
-        audioPaths.push(path);
+        audioPaths.push(recordMediaPath(path));
         return Buffer.from('audio');
       },
       generateScreenshot: async (path) => {
-        imagePaths.push(path);
+        imagePaths.push(recordMediaPath(path));
         return Buffer.from('image');
       },
       generateAnimatedImage: async () => null,

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

@@ -2,6 +2,7 @@ import assert from 'node:assert/strict';
 import test from 'node:test';
 
 import { CardCreationService } from './card-creation';
+import type { MediaInput } from '../media-generator';
 import type { AnkiConnectConfig } from '../types/anki';
 
 test('CardCreationService counts locally created sentence cards', async () => {
@@ -287,6 +288,8 @@ test('CardCreationService keeps updating after recordCardsMinedCallback throws',
 test('CardCreationService uses stream-open-filename for remote media generation', async () => {
   const audioPaths: string[] = [];
   const imagePaths: string[] = [];
+  const recordMediaPath = (mediaInput: MediaInput): string =>
+    typeof mediaInput === 'string' ? mediaInput : mediaInput.path;
   const edlSource = [
     'edl://!new_stream;!no_clip;!no_chapters;%70%https://audio.example/videoplayback?mime=audio%2Fwebm',
     '!new_stream;!no_clip;!no_chapters;%69%https://video.example/videoplayback?mime=video%2Fmp4',
@@ -345,11 +348,11 @@ test('CardCreationService uses stream-open-filename for remote media generation'
     },
     mediaGenerator: {
       generateAudio: async (path) => {
-        audioPaths.push(path);
+        audioPaths.push(recordMediaPath(path));
         return Buffer.from('audio');
       },
       generateScreenshot: async (path) => {
-        imagePaths.push(path);
+        imagePaths.push(recordMediaPath(path));
         return Buffer.from('image');
       },
       generateAnimatedImage: async () => null,
@@ -398,6 +401,259 @@ test('CardCreationService uses stream-open-filename for remote media generation'
   assert.deepEqual(imagePaths, ['https://video.example/videoplayback?mime=video%2Fmp4']);
 });
 
+test('CardCreationService does not use mpv stream indexes for ready cached YouTube media', async () => {
+  const audioCalls: Array<{ path: string; audioStreamIndex?: number }> = [];
+
+  const service = new CardCreationService({
+    getConfig: () =>
+      ({
+        deck: 'Mining',
+        fields: {
+          sentence: 'Sentence',
+          audio: 'SentenceAudio',
+          image: 'Picture',
+        },
+        media: {
+          generateAudio: true,
+          generateImage: false,
+          imageFormat: 'jpg',
+        },
+        behavior: {},
+        ai: false,
+      }) as AnkiConnectConfig,
+    getAiConfig: () => ({}),
+    getTimingTracker: () => ({}) as never,
+    getMpvClient: () =>
+      ({
+        currentVideoPath: 'https://www.youtube.com/watch?v=abc123',
+        currentSubText: '字幕',
+        currentSubStart: 10,
+        currentSubEnd: 12,
+        currentTimePos: 11,
+        currentAudioStreamIndex: 0,
+      }) as never,
+    getCachedMediaPath: async () => '/tmp/subminer-youtube-media-cache/media.mkv',
+    shouldRequireRemoteMediaCache: () => true,
+    client: {
+      addNote: async () => 42,
+      addTags: async () => undefined,
+      notesInfo: async () => [
+        {
+          noteId: 42,
+          fields: {
+            Sentence: { value: '' },
+            SentenceAudio: { value: '' },
+            Picture: { value: '' },
+          },
+        },
+      ],
+      updateNoteFields: async () => undefined,
+      storeMediaFile: async () => undefined,
+      findNotes: async () => [],
+      retrieveMediaFile: async () => '',
+    },
+    mediaGenerator: {
+      generateAudio: async (path, _startTime, _endTime, _padding, audioStreamIndex) => {
+        audioCalls.push({ path: typeof path === 'string' ? path : path.path, audioStreamIndex });
+        return Buffer.from('audio');
+      },
+      generateScreenshot: async () => null,
+      generateAnimatedImage: async () => null,
+    },
+    showOsdNotification: () => undefined,
+    showUpdateResult: () => undefined,
+    showStatusNotification: () => undefined,
+    showNotification: async () => undefined,
+    beginUpdateProgress: () => undefined,
+    endUpdateProgress: () => undefined,
+    withUpdateProgress: async (_message, action) => action(),
+    resolveConfiguredFieldName: (noteInfo, preferredName) => {
+      if (!preferredName) return null;
+      return Object.keys(noteInfo.fields).find((field) => field === preferredName) ?? null;
+    },
+    resolveNoteFieldName: (noteInfo, preferredName) => {
+      if (!preferredName) return null;
+      return Object.keys(noteInfo.fields).find((field) => field === preferredName) ?? null;
+    },
+    getAnimatedImageLeadInSeconds: async () => 0,
+    extractFields: () => ({}),
+    processSentence: (sentence) => sentence,
+    setCardTypeFields: () => undefined,
+    mergeFieldValue: (_existing, newValue) => newValue,
+    formatMiscInfoPattern: () => '',
+    getEffectiveSentenceCardConfig: () => ({
+      model: 'Sentence',
+      sentenceField: 'Sentence',
+      audioField: 'SentenceAudio',
+      lapisEnabled: false,
+      kikuEnabled: false,
+      kikuFieldGrouping: 'disabled',
+      kikuDeleteDuplicateInAuto: false,
+    }),
+    getFallbackDurationSeconds: () => 10,
+    appendKnownWordsFromNoteInfo: () => undefined,
+    isUpdateInProgress: () => false,
+    setUpdateInProgress: () => undefined,
+    trackLastAddedNoteId: () => undefined,
+  });
+
+  const created = await service.createSentenceCard('テスト', 10, 12);
+
+  assert.equal(created, true);
+  assert.deepEqual(audioCalls, [
+    {
+      path: '/tmp/subminer-youtube-media-cache/media.mkv',
+      audioStreamIndex: undefined,
+    },
+  ]);
+});
+
+test('CardCreationService queues YouTube media when required cache is not ready', async () => {
+  const mediaCalls: string[] = [];
+  const updates: Array<{ noteId: number; fields: Record<string, string> }> = [];
+  const queuedUpdates: Array<{
+    sourceUrl: string;
+    noteId: number;
+    startTime: number;
+    endTime: number;
+    label: string | number;
+    audioFieldName?: string;
+    imageFieldName?: string;
+    miscInfoFieldName?: string;
+    generateAudio: boolean;
+    generateImage: boolean;
+  }> = [];
+  let streamRequests = 0;
+
+  const service = new CardCreationService({
+    getConfig: () =>
+      ({
+        deck: 'Mining',
+        fields: {
+          sentence: 'Sentence',
+          audio: 'SentenceAudio',
+          image: 'Picture',
+          miscInfo: 'MiscInfo',
+        },
+        media: {
+          generateAudio: true,
+          generateImage: true,
+          imageFormat: 'jpg',
+        },
+        behavior: {},
+        ai: false,
+      }) as AnkiConnectConfig,
+    getAiConfig: () => ({}),
+    getTimingTracker: () => ({}) as never,
+    getMpvClient: () =>
+      ({
+        currentVideoPath: 'https://www.youtube.com/watch?v=abc123',
+        currentSubText: '字幕',
+        currentSubStart: 10,
+        currentSubEnd: 12,
+        currentTimePos: 11,
+        currentAudioStreamIndex: 2,
+        requestProperty: async () => {
+          streamRequests += 1;
+          return 'https://rr1---sn.example.googlevideo.com/videoplayback?id=123';
+        },
+      }) as never,
+    getCachedMediaPath: async () => null,
+    shouldRequireRemoteMediaCache: () => true,
+    queuePendingYoutubeMediaUpdate: (job) => {
+      queuedUpdates.push(job);
+    },
+    client: {
+      addNote: async () => 42,
+      addTags: async () => undefined,
+      notesInfo: async () => [
+        {
+          noteId: 42,
+          fields: {
+            Sentence: { value: '' },
+            SentenceAudio: { value: '' },
+            Picture: { value: '' },
+            MiscInfo: { value: '' },
+          },
+        },
+      ],
+      updateNoteFields: async (noteId, fields) => {
+        updates.push({ noteId, fields });
+      },
+      storeMediaFile: async () => undefined,
+      findNotes: async () => [],
+      retrieveMediaFile: async () => '',
+    },
+    mediaGenerator: {
+      generateAudio: async () => {
+        mediaCalls.push('audio');
+        return Buffer.from('audio');
+      },
+      generateScreenshot: async () => {
+        mediaCalls.push('image');
+        return Buffer.from('image');
+      },
+      generateAnimatedImage: async () => null,
+    },
+    showOsdNotification: () => undefined,
+    showUpdateResult: () => undefined,
+    showStatusNotification: () => undefined,
+    showNotification: async () => undefined,
+    beginUpdateProgress: () => undefined,
+    endUpdateProgress: () => undefined,
+    withUpdateProgress: async (_message, action) => action(),
+    resolveConfiguredFieldName: (noteInfo, preferredName) => {
+      if (!preferredName) return null;
+      return Object.keys(noteInfo.fields).find((field) => field === preferredName) ?? null;
+    },
+    resolveNoteFieldName: (noteInfo, preferredName) => {
+      if (!preferredName) return null;
+      return Object.keys(noteInfo.fields).find((field) => field === preferredName) ?? null;
+    },
+    getAnimatedImageLeadInSeconds: async () => 0,
+    extractFields: () => ({}),
+    processSentence: (sentence) => sentence,
+    setCardTypeFields: () => undefined,
+    mergeFieldValue: (_existing, newValue) => newValue,
+    formatMiscInfoPattern: () => '',
+    getEffectiveSentenceCardConfig: () => ({
+      model: 'Sentence',
+      sentenceField: 'Sentence',
+      audioField: 'SentenceAudio',
+      lapisEnabled: false,
+      kikuEnabled: false,
+      kikuFieldGrouping: 'disabled',
+      kikuDeleteDuplicateInAuto: false,
+    }),
+    getFallbackDurationSeconds: () => 10,
+    appendKnownWordsFromNoteInfo: () => undefined,
+    isUpdateInProgress: () => false,
+    setUpdateInProgress: () => undefined,
+    trackLastAddedNoteId: () => undefined,
+  });
+
+  const created = await service.createSentenceCard('テスト', 10, 12);
+
+  assert.equal(created, true);
+  assert.equal(streamRequests, 0);
+  assert.deepEqual(mediaCalls, []);
+  assert.deepEqual(queuedUpdates, [
+    {
+      sourceUrl: 'https://www.youtube.com/watch?v=abc123',
+      noteId: 42,
+      startTime: 10,
+      endTime: 12,
+      label: 'テスト',
+      audioFieldName: 'SentenceAudio',
+      imageFieldName: 'Picture',
+      miscInfoFieldName: 'MiscInfo',
+      generateAudio: true,
+      generateImage: true,
+    },
+  ]);
+  assert.deepEqual(updates, []);
+});
+
 test('CardCreationService tracks pre-add duplicate note ids for kiku sentence cards', async () => {
   const trackedDuplicates: Array<{ noteId: number; duplicateNoteIds: number[] }> = [];
   const duplicateLookupExpressions: string[] = [];

src/anki-integration/card-creation.ts

@@ -5,15 +5,29 @@ import {
 } from '../anki-field-config';
 import { AnkiConnectConfig } from '../types/anki';
 import { createLogger } from '../logger';
+import type { MediaInput } from '../media-input';
 import { SubtitleTimingTracker } from '../subtitle-timing-tracker';
 import { AiConfig } from '../types/integrations';
 import { MpvClient } from '../types/runtime';
 import { resolveSentenceBackText } from './ai';
-import { resolveMediaGenerationInputPath } from './media-source';
+import {
+  resolveMediaGenerationInput,
+  resolveAudioStreamIndexForMediaGeneration,
+  type MediaGenerationInputResolverOptions,
+} from './media-source';
 import { shouldMarkWordAndSentenceCard } from './note-field-utils';
+import type { PendingYoutubeMediaUpdate } from './pending-youtube-media';
 
 const log = createLogger('anki').child('integration.card-creation');
 
+function shouldGenerateAudio(config: AnkiConnectConfig): boolean {
+  return config.media?.generateAudio !== false;
+}
+
+function shouldGenerateImage(config: AnkiConnectConfig): boolean {
+  return config.media?.generateImage !== false;
+}
+
 export interface CardCreationNoteInfo {
   noteId: number;
   fields: Record<string, { value: string }>;
@@ -38,14 +52,14 @@ interface CardCreationClient {
 
 interface CardCreationMediaGenerator {
   generateAudio(
-    path: string,
+    path: MediaInput,
     startTime: number,
     endTime: number,
     audioPadding?: number,
     audioStreamIndex?: number,
   ): Promise<Buffer | null>;
   generateScreenshot(
-    path: string,
+    path: MediaInput,
     timestamp: number,
     options: {
       format: 'jpg' | 'png' | 'webp';
@@ -55,7 +69,7 @@ interface CardCreationMediaGenerator {
     },
   ): Promise<Buffer | null>;
   generateAnimatedImage(
-    path: string,
+    path: MediaInput,
     startTime: number,
     endTime: number,
     audioPadding?: number,
@@ -74,6 +88,9 @@ interface CardCreationDeps {
   getAiConfig: () => AiConfig;
   getTimingTracker: () => SubtitleTimingTracker;
   getMpvClient: () => MpvClient;
+  getCachedMediaPath?: MediaGenerationInputResolverOptions['getCachedMediaPath'];
+  shouldRequireRemoteMediaCache?: () => boolean;
+  queuePendingYoutubeMediaUpdate?: (job: PendingYoutubeMediaUpdate) => void;
   getDeck?: () => string | undefined;
   client: CardCreationClient;
   mediaGenerator: CardCreationMediaGenerator;
@@ -121,6 +138,19 @@ interface CardCreationDeps {
 export class CardCreationService {
   constructor(private readonly deps: CardCreationDeps) {}
 
+  private getMediaResolverOptions(): MediaGenerationInputResolverOptions {
+    const options: MediaGenerationInputResolverOptions = {
+      logDebug: (message) => log.debug(message),
+    };
+    if (this.deps.getCachedMediaPath) {
+      options.getCachedMediaPath = this.deps.getCachedMediaPath;
+    }
+    if (this.deps.shouldRequireRemoteMediaCache?.()) {
+      options.remoteCacheMode = 'required';
+    }
+    return options;
+  }
+
   private getConfiguredAnkiTags(): string[] {
     const tags = this.deps.getConfig().tags;
     if (!Array.isArray(tags)) {
@@ -246,14 +276,18 @@ export class CardCreationService {
           `Clipboard update: timing range ${rangeStart.toFixed(2)}s - ${rangeEnd.toFixed(2)}s`,
         );
 
-        const audioSourcePath = this.deps.getConfig().media?.generateAudio
-          ? await resolveMediaGenerationInputPath(mpvClient, 'audio')
+        const config = this.deps.getConfig();
+        const generateAudio = shouldGenerateAudio(config);
+        const generateImage = shouldGenerateImage(config);
+        const mediaResolverOptions = this.getMediaResolverOptions();
+        const audioSourcePath = generateAudio
+          ? await resolveMediaGenerationInput(mpvClient, 'audio', mediaResolverOptions)
           : null;
-        const videoPath = this.deps.getConfig().media?.generateImage
-          ? await resolveMediaGenerationInputPath(mpvClient, 'video')
+        const videoPath = generateImage
+          ? await resolveMediaGenerationInput(mpvClient, 'video', mediaResolverOptions)
           : null;
 
-        if (this.deps.getConfig().media?.generateAudio) {
+        if (generateAudio) {
           try {
             const audioFilename = this.generateAudioFilename();
             const audioBuffer = audioSourcePath
@@ -281,7 +315,7 @@ export class CardCreationService {
           }
         }
 
-        if (this.deps.getConfig().media?.generateImage) {
+        if (generateImage) {
           try {
             const animatedLeadInSeconds = await this.deps.getAnimatedImageLeadInSeconds(noteInfo);
             const imageFilename = this.generateImageFilename();
@@ -441,7 +475,7 @@ export class CardCreationService {
           errors.push('audio');
         }
 
-        if (this.deps.getConfig().media?.generateImage) {
+        if (shouldGenerateImage(this.deps.getConfig())) {
           try {
             const animatedLeadInSeconds = await this.deps.getAnimatedImageLeadInSeconds(noteInfo);
             const imageFilename = this.generateImageFilename();
@@ -522,9 +556,23 @@ export class CardCreationService {
 
     try {
       return await this.deps.withUpdateProgress('Creating sentence card', async () => {
-        const videoPath = await resolveMediaGenerationInputPath(mpvClient, 'video');
-        const audioSourcePath = await resolveMediaGenerationInputPath(mpvClient, 'audio');
-        if (!videoPath) {
+        const config = this.deps.getConfig();
+        const generateAudio = shouldGenerateAudio(config);
+        const generateImage = shouldGenerateImage(config);
+        const mediaResolverOptions = this.getMediaResolverOptions();
+        const videoPath = generateImage
+          ? await resolveMediaGenerationInput(mpvClient, 'video', mediaResolverOptions)
+          : null;
+        const audioSourcePath = generateAudio
+          ? await resolveMediaGenerationInput(mpvClient, 'audio', mediaResolverOptions)
+          : null;
+        const missingRequestedMediaInput =
+          (generateImage && !videoPath) || (generateAudio && !audioSourcePath);
+        const shouldQueuePendingYoutubeMedia =
+          missingRequestedMediaInput &&
+          this.deps.shouldRequireRemoteMediaCache?.() === true &&
+          typeof this.deps.queuePendingYoutubeMediaUpdate === 'function';
+        if (missingRequestedMediaInput && !shouldQueuePendingYoutubeMedia) {
           this.deps.showOsdNotification('No video loaded');
           return false;
         }
@@ -534,13 +582,13 @@ export class CardCreationService {
 
         const sentenceField = sentenceCardConfig.sentenceField;
         const audioFieldName = sentenceCardConfig.audioField || 'SentenceAudio';
-        const translationField = this.deps.getConfig().fields?.translation || 'SelectionText';
+        const translationField = config.fields?.translation || 'SelectionText';
         let resolvedMiscInfoField: string | null = null;
         let resolvedSentenceAudioField: string = audioFieldName;
 
         fields[sentenceField] = sentence;
 
-        const ankiAiConfig = this.deps.getConfig().ai;
+        const ankiAiConfig = config.ai;
         const ankiAiEnabled =
           typeof ankiAiConfig === 'object' && ankiAiConfig !== null
             ? ankiAiConfig.enabled === true
@@ -563,7 +611,7 @@ export class CardCreationService {
 
         if (sentenceCardConfig.lapisEnabled || sentenceCardConfig.kikuEnabled) {
           fields.IsSentenceCard = 'x';
-          fields[getConfiguredWordFieldName(this.deps.getConfig())] = sentence;
+          fields[getConfiguredWordFieldName(config)] = sentence;
         }
 
         const pendingNoteInfo = this.createPendingNoteInfo(fields);
@@ -572,7 +620,7 @@ export class CardCreationService {
         );
         const pendingExpressionText = getPreferredWordValueFromExtractedFields(
           pendingNoteFields,
-          this.deps.getConfig(),
+          config,
         ).trim();
         let duplicateNoteIds: number[] = [];
         if (
@@ -590,7 +638,7 @@ export class CardCreationService {
           }
         }
 
-        const deck = this.deps.getConfig().deck || 'Default';
+        const deck = config.deck || 'Default';
         let noteId: number;
         try {
           noteId = await this.deps.client.addNote(
@@ -636,7 +684,7 @@ export class CardCreationService {
               this.deps.resolveNoteFieldName(createdNoteInfo, audioFieldName) || audioFieldName;
             resolvedMiscInfoField = this.deps.resolveConfiguredFieldName(
               createdNoteInfo,
-              this.deps.getConfig().fields?.miscInfo,
+              config.fields?.miscInfo,
             );
 
             const cardTypeFields: Record<string, string> = {};
@@ -654,8 +702,31 @@ export class CardCreationService {
           errors.push('card type fields');
         }
 
+        const label = sentence.length > 30 ? sentence.substring(0, 30) + '...' : sentence;
+        if (shouldQueuePendingYoutubeMedia) {
+          this.deps.queuePendingYoutubeMediaUpdate?.({
+            sourceUrl: mpvClient.currentVideoPath,
+            noteId,
+            startTime,
+            endTime,
+            label,
+            audioFieldName: resolvedSentenceAudioField,
+            imageFieldName: config.fields?.image,
+            miscInfoFieldName: resolvedMiscInfoField ?? undefined,
+            generateAudio,
+            generateImage,
+          });
+          await this.deps.showNotification(noteId, label, 'media queued');
+          return true;
+        }
+
+        if (missingRequestedMediaInput) {
+          return false;
+        }
+
         const mediaFields: Record<string, string> = {};
 
+        if (generateAudio) {
           try {
             const audioFilename = this.generateAudioFilename();
             const audioBuffer = audioSourcePath
@@ -672,12 +743,14 @@ export class CardCreationService {
             log.error('Failed to generate sentence audio:', (error as Error).message);
             errors.push('audio');
           }
+        }
 
+        if (generateImage) {
           try {
             const imageFilename = this.generateImageFilename();
-          const imageBuffer = await this.generateImageBuffer(videoPath, startTime, endTime);
+            const imageBuffer = await this.generateImageBuffer(videoPath!, startTime, endTime);
 
-          const imageField = this.deps.getConfig().fields?.image;
+            const imageField = config.fields?.image;
             if (imageBuffer && imageField) {
               await this.deps.client.storeMediaFile(imageFilename, imageBuffer);
               mediaFields[imageField] = `<img src="${imageFilename}">`;
@@ -687,8 +760,9 @@ export class CardCreationService {
             log.error('Failed to generate sentence image:', (error as Error).message);
             errors.push('image');
           }
+        }
 
-        if (this.deps.getConfig().fields?.miscInfo) {
+        if (config.fields?.miscInfo) {
           const miscInfo = this.deps.formatMiscInfoPattern(miscInfoFilename || '', startTime);
           if (miscInfo && resolvedMiscInfoField) {
             mediaFields[resolvedMiscInfoField] = miscInfo;
@@ -704,7 +778,6 @@ export class CardCreationService {
           }
         }
 
-        const label = sentence.length > 30 ? sentence.substring(0, 30) + '...' : sentence;
         const errorSuffix = errors.length > 0 ? `${errors.join(', ')} failed` : undefined;
         await this.deps.showNotification(noteId, label, errorSuffix);
         return true;
@@ -740,7 +813,7 @@ export class CardCreationService {
   }
 
   private async mediaGenerateAudio(
-    videoPath: string,
+    videoPath: MediaInput,
     startTime: number,
     endTime: number,
   ): Promise<Buffer | null> {
@@ -754,12 +827,15 @@ export class CardCreationService {
       startTime,
       endTime,
       this.deps.getConfig().media?.audioPadding,
+      resolveAudioStreamIndexForMediaGeneration(
+        videoPath,
         mpvClient.currentAudioStreamIndex ?? undefined,
+      ),
     );
   }
 
   private async generateImageBuffer(
-    videoPath: string,
+    videoPath: MediaInput,
     startTime: number,
     endTime: number,
     animatedLeadInSeconds = 0,

src/anki-integration/media-source.test.ts

@@ -1,7 +1,33 @@
 import assert from 'node:assert/strict';
 import test from 'node:test';
 
-import { resolveMediaGenerationInputPath } from './media-source';
+import * as mediaSource from './media-source';
+
+const { resolveMediaGenerationInputPath } = mediaSource;
+
+type StructuredMediaInput = {
+  path: string;
+  source: string;
+  singleResolvedStream: boolean;
+  inputOptions?: {
+    reconnect?: boolean;
+    userAgent?: string;
+    headers?: Record<string, string>;
+  };
+};
+
+type StructuredMediaResolver = (
+  mpvClient: Parameters<typeof resolveMediaGenerationInputPath>[0],
+  kind?: Parameters<typeof resolveMediaGenerationInputPath>[1],
+  options?: {
+    getCachedMediaPath?: (
+      currentVideoPath: string,
+      kind: Parameters<typeof resolveMediaGenerationInputPath>[1],
+    ) => Promise<string | null>;
+    remoteCacheMode?: 'optional' | 'required';
+    logDebug?: (message: string) => void;
+  },
+) => Promise<StructuredMediaInput | null>;
 
 test('resolveMediaGenerationInputPath keeps local file paths', async () => {
   const result = await resolveMediaGenerationInputPath({
@@ -62,3 +88,211 @@ test('resolveMediaGenerationInputPath falls back to currentVideoPath when stream
 
   assert.equal(result, 'https://www.youtube.com/watch?v=abc123');
 });
+
+test('resolveMediaGenerationInput returns single-stream metadata for mpv EDL URLs', async () => {
+  const resolver = (
+    mediaSource as typeof mediaSource & {
+      resolveMediaGenerationInput?: StructuredMediaResolver;
+    }
+  ).resolveMediaGenerationInput;
+  assert.equal(typeof resolver, 'function');
+
+  const edlSource = [
+    'edl://!new_stream;!no_clip;!no_chapters;%70%https://audio.example/videoplayback?mime=audio%2Fwebm',
+    '!new_stream;!no_clip;!no_chapters;%69%https://video.example/videoplayback?mime=video%2Fmp4',
+  ].join(';');
+
+  const result = await resolver!(
+    {
+      currentVideoPath: 'https://www.youtube.com/watch?v=abc123',
+      requestProperty: async (name: string) => {
+        if (name === 'stream-open-filename') return edlSource;
+        if (name === 'user-agent') return 'Mozilla/5.0';
+        if (name === 'http-header-fields') {
+          return ['Cookie: SID=secret', 'Referer: https://www.youtube.com/', 'X-Test: ok'];
+        }
+        return null;
+      },
+    },
+    'audio',
+  );
+
+  assert.equal(result?.path, 'https://audio.example/videoplayback?mime=audio%2Fwebm');
+  assert.equal(result?.singleResolvedStream, true);
+  assert.equal(result?.inputOptions?.reconnect, true);
+  assert.equal(result?.inputOptions?.userAgent, 'Mozilla/5.0');
+  assert.deepEqual(result?.inputOptions?.headers, {
+    Referer: 'https://www.youtube.com/',
+    'X-Test': 'ok',
+  });
+});
+
+test('resolveMediaGenerationInput reads file-local mpv request options', async () => {
+  const resolver = (
+    mediaSource as typeof mediaSource & {
+      resolveMediaGenerationInput?: StructuredMediaResolver;
+    }
+  ).resolveMediaGenerationInput;
+  assert.equal(typeof resolver, 'function');
+
+  const result = await resolver!(
+    {
+      currentVideoPath: 'https://www.youtube.com/watch?v=abc123',
+      requestProperty: async (name: string) => {
+        if (name === 'stream-open-filename') {
+          return 'https://rr1---sn.example.googlevideo.com/videoplayback?id=123';
+        }
+        if (name === 'file-local-options/user-agent') return 'SubMiner Test Agent';
+        if (name === 'options/http-header-fields') return ['X-Shared: ok'];
+        if (name === 'file-local-options/http-header-fields') {
+          return ['Cookie: SID=secret', 'Referer: https://m.youtube.com/', 'X-Local: yes'];
+        }
+        return null;
+      },
+    },
+    'video',
+  );
+
+  assert.equal(result?.path, 'https://rr1---sn.example.googlevideo.com/videoplayback?id=123');
+  assert.equal(result?.singleResolvedStream, true);
+  assert.equal(result?.inputOptions?.userAgent, 'SubMiner Test Agent');
+  assert.deepEqual(result?.inputOptions?.headers, {
+    'X-Shared': 'ok',
+    Referer: 'https://m.youtube.com/',
+    'X-Local': 'yes',
+    Origin: 'https://www.youtube.com',
+  });
+});
+
+test('resolveMediaGenerationInput prefers a ready cached media file for YouTube extraction', async () => {
+  const resolver = (
+    mediaSource as typeof mediaSource & {
+      resolveMediaGenerationInput?: StructuredMediaResolver;
+    }
+  ).resolveMediaGenerationInput;
+  assert.equal(typeof resolver, 'function');
+
+  const result = await resolver!(
+    {
+      currentVideoPath: 'https://www.youtube.com/watch?v=abc123',
+      requestProperty: async () => 'https://rr1---sn.example.googlevideo.com/videoplayback?id=123',
+    },
+    'video',
+    {
+      getCachedMediaPath: async () => '/tmp/subminer-youtube-media-cache/abc123/media.mkv',
+    },
+  );
+
+  assert.equal(result?.path, '/tmp/subminer-youtube-media-cache/abc123/media.mkv');
+  assert.equal(result?.source, 'youtube-cache');
+  assert.equal(result?.singleResolvedStream, false);
+  assert.equal(result?.inputOptions, undefined);
+});
+
+test('resolveMediaGenerationInput debug-logs sanitized YouTube cache hits', async () => {
+  const resolver = (
+    mediaSource as typeof mediaSource & {
+      resolveMediaGenerationInput?: StructuredMediaResolver;
+    }
+  ).resolveMediaGenerationInput;
+  assert.equal(typeof resolver, 'function');
+  const logs: string[] = [];
+
+  const result = await resolver!(
+    {
+      currentVideoPath: 'https://www.youtube.com/watch?v=abc123&signature=secret',
+      requestProperty: async () => 'https://rr1---sn.example.googlevideo.com/videoplayback?id=123',
+    },
+    'video',
+    {
+      getCachedMediaPath: async () => '/tmp/subminer-youtube-media-cache/abc123/media.mkv',
+      logDebug: (message) => logs.push(message),
+    },
+  );
+
+  assert.equal(result?.source, 'youtube-cache');
+  assert.match(logs.join('\n'), /kind=video source=youtube-cache/);
+  assert.match(
+    logs.join('\n'),
+    /input=local:\/tmp\/subminer-youtube-media-cache\/abc123\/media\.mkv/,
+  );
+  assert.match(logs.join('\n'), /current=remote:www\.youtube\.com/);
+  assert.doesNotMatch(logs.join('\n'), /signature=secret|videoplayback/);
+});
+
+test('resolveMediaGenerationInput does not fall back to direct remote streams when cache is required', async () => {
+  const resolver = (
+    mediaSource as typeof mediaSource & {
+      resolveMediaGenerationInput?: StructuredMediaResolver;
+    }
+  ).resolveMediaGenerationInput;
+  assert.equal(typeof resolver, 'function');
+
+  const result = await resolver!(
+    {
+      currentVideoPath: 'https://www.youtube.com/watch?v=abc123',
+      requestProperty: async () => 'https://rr1---sn.example.googlevideo.com/videoplayback?id=123',
+    },
+    'video',
+    {
+      getCachedMediaPath: async () => null,
+      remoteCacheMode: 'required',
+    },
+  );
+
+  assert.equal(result, null);
+});
+
+test('resolveMediaGenerationInput falls back when optional cache lookup fails', async () => {
+  const resolver = (
+    mediaSource as typeof mediaSource & {
+      resolveMediaGenerationInput?: StructuredMediaResolver;
+    }
+  ).resolveMediaGenerationInput;
+  assert.equal(typeof resolver, 'function');
+
+  const result = await resolver!(
+    {
+      currentVideoPath: 'https://www.youtube.com/watch?v=abc123',
+      requestProperty: async () => 'https://rr1---sn.example.googlevideo.com/videoplayback?id=123',
+    },
+    'video',
+    {
+      getCachedMediaPath: async () => {
+        throw new Error('cache unavailable');
+      },
+    },
+  );
+
+  assert.equal(result?.path, 'https://rr1---sn.example.googlevideo.com/videoplayback?id=123');
+  assert.equal(result?.source, 'stream-open-filename');
+});
+
+test('resolveMediaGenerationInput debug-logs sanitized required-cache misses', async () => {
+  const resolver = (
+    mediaSource as typeof mediaSource & {
+      resolveMediaGenerationInput?: StructuredMediaResolver;
+    }
+  ).resolveMediaGenerationInput;
+  assert.equal(typeof resolver, 'function');
+  const logs: string[] = [];
+
+  const result = await resolver!(
+    {
+      currentVideoPath: 'https://www.youtube.com/watch?v=abc123&signature=secret',
+      requestProperty: async () => 'https://rr1---sn.example.googlevideo.com/videoplayback?id=123',
+    },
+    'video',
+    {
+      getCachedMediaPath: async () => null,
+      remoteCacheMode: 'required',
+      logDebug: (message) => logs.push(message),
+    },
+  );
+
+  assert.equal(result, null);
+  assert.match(logs.join('\n'), /kind=video source=cache-miss/);
+  assert.match(logs.join('\n'), /mode=required/);
+  assert.match(logs.join('\n'), /current=remote:www\.youtube\.com/);
+  assert.doesNotMatch(logs.join('\n'), /signature=secret|videoplayback|googlevideo/);
+});

src/anki-integration/media-source.ts

@@ -1,7 +1,57 @@
 import { isRemoteMediaPath } from '../jimaku/utils';
+import type { MediaInput, MediaInputOptions } from '../media-input';
 import type { MpvClient } from '../types/runtime';
 
 export type MediaGenerationKind = 'audio' | 'video';
+export type MediaGenerationInputSource =
+  | 'current-path'
+  | 'stream-open-filename'
+  | 'edl-stream'
+  | 'youtube-cache';
+
+export interface ResolvedMediaGenerationInput {
+  path: string;
+  kind: MediaGenerationKind;
+  source: MediaGenerationInputSource;
+  singleResolvedStream: boolean;
+  inputOptions?: MediaInputOptions;
+}
+
+export interface MediaGenerationInputResolverOptions {
+  getCachedMediaPath?: (
+    currentVideoPath: string,
+    kind: MediaGenerationKind,
+  ) => Promise<string | null>;
+  remoteCacheMode?: 'optional' | 'required';
+  logDebug?: (message: string) => void;
+}
+
+export function resolveAudioStreamIndexForMediaGeneration(
+  input: MediaInput,
+  audioStreamIndex: number | null | undefined,
+): number | undefined {
+  if (typeof input === 'object' && 'source' in input && input.source === 'youtube-cache') {
+    return undefined;
+  }
+  return audioStreamIndex ?? undefined;
+}
+
+const BLOCKED_HTTP_HEADER_NAMES = new Set(['authorization', 'cookie', 'proxy-authorization']);
+const HTTP_HEADER_FIELD_PROPERTY_NAMES = [
+  'http-header-fields',
+  'options/http-header-fields',
+  'file-local-options/http-header-fields',
+] as const;
+const USER_AGENT_PROPERTY_NAMES = [
+  'file-local-options/user-agent',
+  'options/user-agent',
+  'user-agent',
+] as const;
+const REFERRER_PROPERTY_NAMES = [
+  'file-local-options/referrer',
+  'options/referrer',
+  'referrer',
+] as const;
 
 function trimToNonEmptyString(value: unknown): string | null {
   if (typeof value !== 'string') {
@@ -11,6 +61,17 @@ function trimToNonEmptyString(value: unknown): string | null {
   return trimmed.length > 0 ? trimmed : null;
 }
 
+function normalizeHeaderName(value: string): string | null {
+  const trimmed = value.trim();
+  if (!/^[A-Za-z0-9!#$%&'*+.^_`|~-]+$/.test(trimmed)) {
+    return null;
+  }
+  if (BLOCKED_HTTP_HEADER_NAMES.has(trimmed.toLowerCase())) {
+    return null;
+  }
+  return trimmed;
+}
+
 function extractUrlsFromMpvEdlSource(source: string): string[] {
   const matches = source.matchAll(/%\d+%(https?:\/\/.*?)(?=;!new_stream|;!global_tags|$)/gms);
   return [...matches]
@@ -53,6 +114,317 @@ function resolvePreferredUrlFromMpvEdlSource(
   return kind === 'audio' ? (urls[0] ?? null) : (urls[urls.length - 1] ?? null);
 }
 
+function getHostname(value: string): string | null {
+  try {
+    return new URL(value).hostname.toLowerCase();
+  } catch {
+    return null;
+  }
+}
+
+function matchesHost(hostname: string, expectedHost: string): boolean {
+  return hostname === expectedHost || hostname.endsWith(`.${expectedHost}`);
+}
+
+function isGoogleVideoMediaPath(value: string): boolean {
+  const host = getHostname(value);
+  return Boolean(host && matchesHost(host, 'googlevideo.com'));
+}
+
+function describeMediaPathForDebugLog(value: string): string {
+  try {
+    const url = new URL(value);
+    if (url.protocol === 'http:' || url.protocol === 'https:') {
+      return `remote:${url.hostname.toLowerCase() || 'unknown'}`;
+    }
+    return `${url.protocol.replace(/:$/, '')}:`;
+  } catch {
+    // Not a URL; treat as a local file path below.
+  }
+
+  if (value.startsWith('edl://')) {
+    return 'edl:';
+  }
+
+  return `local:${value}`;
+}
+
+function logMediaResolutionDebug(
+  options: MediaGenerationInputResolverOptions,
+  message: string,
+): void {
+  if (!options.logDebug) {
+    return;
+  }
+  try {
+    options.logDebug(`[media-source] ${message}`);
+  } catch {
+    // Debug logging should not affect media generation.
+  }
+}
+
+function logResolvedMediaGenerationInput(
+  options: MediaGenerationInputResolverOptions,
+  currentVideoPath: string,
+  result: ResolvedMediaGenerationInput,
+): void {
+  logMediaResolutionDebug(
+    options,
+    [
+      `kind=${result.kind}`,
+      `source=${result.source}`,
+      `input=${describeMediaPathForDebugLog(result.path)}`,
+      `current=${describeMediaPathForDebugLog(currentVideoPath)}`,
+      `singleResolvedStream=${result.singleResolvedStream}`,
+    ].join(' '),
+  );
+}
+
+function logMediaGenerationInputMiss(
+  options: MediaGenerationInputResolverOptions,
+  kind: MediaGenerationKind,
+  currentVideoPath: string,
+  reason: string,
+): void {
+  logMediaResolutionDebug(
+    options,
+    [
+      `kind=${kind}`,
+      'source=cache-miss',
+      `reason=${reason}`,
+      `mode=${options.remoteCacheMode ?? 'optional'}`,
+      `current=${describeMediaPathForDebugLog(currentVideoPath)}`,
+    ].join(' '),
+  );
+}
+
+function setHeaderIfMissing(headers: Record<string, string>, name: string, value: string): void {
+  const lowerName = name.toLowerCase();
+  if (!Object.keys(headers).some((existing) => existing.toLowerCase() === lowerName)) {
+    headers[name] = value;
+  }
+}
+
+function parseMpvHeaderField(value: string): [string, string] | null {
+  const separatorIndex = value.indexOf(':');
+  if (separatorIndex <= 0) {
+    return null;
+  }
+
+  const name = normalizeHeaderName(value.slice(0, separatorIndex));
+  const headerValue = trimToNonEmptyString(value.slice(separatorIndex + 1));
+  if (!name || !headerValue) {
+    return null;
+  }
+  return [name, headerValue.replace(/[\r\n]+/g, ' ')];
+}
+
+function toHeaderFields(value: unknown): string[] {
+  if (Array.isArray(value)) {
+    return value.filter((entry): entry is string => typeof entry === 'string');
+  }
+  if (typeof value === 'string') {
+    return value
+      .split(/\r?\n/)
+      .map((entry) => entry.trim())
+      .filter((entry) => entry.length > 0);
+  }
+  return [];
+}
+
+async function requestOptionalMpvProperty(
+  mpvClient: Pick<MpvClient, 'requestProperty'>,
+  name: string,
+): Promise<unknown> {
+  if (!mpvClient.requestProperty) {
+    return null;
+  }
+
+  try {
+    return await mpvClient.requestProperty(name);
+  } catch {
+    return null;
+  }
+}
+
+async function requestFirstNonEmptyStringProperty(
+  mpvClient: Pick<MpvClient, 'requestProperty'>,
+  names: readonly string[],
+): Promise<string | null> {
+  for (const name of names) {
+    const value = trimToNonEmptyString(await requestOptionalMpvProperty(mpvClient, name));
+    if (value) {
+      return value;
+    }
+  }
+  return null;
+}
+
+async function resolveRemoteInputOptions(
+  mpvClient: Pick<MpvClient, 'requestProperty'>,
+  resolvedPath: string,
+): Promise<MediaInputOptions | undefined> {
+  if (!isRemoteMediaPath(resolvedPath) || !mpvClient.requestProperty) {
+    return undefined;
+  }
+
+  const headers: Record<string, string> = {};
+  for (const propertyName of HTTP_HEADER_FIELD_PROPERTY_NAMES) {
+    const mpvHeaderFields = toHeaderFields(
+      await requestOptionalMpvProperty(mpvClient, propertyName),
+    );
+    for (const field of mpvHeaderFields) {
+      const parsed = parseMpvHeaderField(field);
+      if (parsed) {
+        headers[parsed[0]] = parsed[1];
+      }
+    }
+  }
+
+  const userAgent = await requestFirstNonEmptyStringProperty(mpvClient, USER_AGENT_PROPERTY_NAMES);
+  const referrer = await requestFirstNonEmptyStringProperty(mpvClient, REFERRER_PROPERTY_NAMES);
+  if (referrer) {
+    setHeaderIfMissing(headers, 'Referer', referrer);
+  }
+  if (isGoogleVideoMediaPath(resolvedPath)) {
+    setHeaderIfMissing(headers, 'Referer', 'https://www.youtube.com/');
+    setHeaderIfMissing(headers, 'Origin', 'https://www.youtube.com');
+  }
+
+  return {
+    reconnect: true,
+    ...(userAgent ? { userAgent } : {}),
+    ...(Object.keys(headers).length > 0 ? { headers } : {}),
+  };
+}
+
+async function toResolvedMediaGenerationInput(
+  mpvClient: Pick<MpvClient, 'requestProperty'>,
+  path: string,
+  kind: MediaGenerationKind,
+  source: MediaGenerationInputSource,
+  singleResolvedStream: boolean,
+): Promise<ResolvedMediaGenerationInput> {
+  const inputOptions = await resolveRemoteInputOptions(mpvClient, path);
+  return {
+    path,
+    kind,
+    source,
+    singleResolvedStream,
+    ...(inputOptions ? { inputOptions } : {}),
+  };
+}
+
+export async function resolveMediaGenerationInput(
+  mpvClient: Pick<MpvClient, 'currentVideoPath' | 'requestProperty'> | null | undefined,
+  kind: MediaGenerationKind = 'video',
+  options: MediaGenerationInputResolverOptions = {},
+): Promise<ResolvedMediaGenerationInput | null> {
+  const currentVideoPath = trimToNonEmptyString(mpvClient?.currentVideoPath);
+  if (!currentVideoPath) {
+    logMediaResolutionDebug(options, `kind=${kind} source=none reason=no-current-video`);
+    return null;
+  }
+
+  if (!isRemoteMediaPath(currentVideoPath)) {
+    const result: ResolvedMediaGenerationInput = {
+      path: currentVideoPath,
+      kind,
+      source: 'current-path',
+      singleResolvedStream: false,
+    };
+    logResolvedMediaGenerationInput(options, currentVideoPath, result);
+    return result;
+  }
+
+  let cachedPath: string | null = null;
+  if (options.getCachedMediaPath) {
+    try {
+      cachedPath = trimToNonEmptyString(await options.getCachedMediaPath(currentVideoPath, kind));
+    } catch {
+      cachedPath = null;
+    }
+  }
+  if (cachedPath) {
+    const result: ResolvedMediaGenerationInput = {
+      path: cachedPath,
+      kind,
+      source: 'youtube-cache',
+      singleResolvedStream: false,
+    };
+    logResolvedMediaGenerationInput(options, currentVideoPath, result);
+    return result;
+  }
+
+  if (options.remoteCacheMode === 'required') {
+    logMediaGenerationInputMiss(options, kind, currentVideoPath, 'required-cache-unavailable');
+    return null;
+  }
+
+  if (!mpvClient?.requestProperty) {
+    const result: ResolvedMediaGenerationInput = {
+      path: currentVideoPath,
+      kind,
+      source: 'current-path',
+      singleResolvedStream: false,
+    };
+    logResolvedMediaGenerationInput(options, currentVideoPath, result);
+    return result;
+  }
+
+  try {
+    const streamOpenFilename = trimToNonEmptyString(
+      await mpvClient.requestProperty('stream-open-filename'),
+    );
+    if (streamOpenFilename?.startsWith('edl://')) {
+      const preferredUrl = resolvePreferredUrlFromMpvEdlSource(streamOpenFilename, kind);
+      if (preferredUrl) {
+        const result = await toResolvedMediaGenerationInput(
+          mpvClient,
+          preferredUrl,
+          kind,
+          'edl-stream',
+          true,
+        );
+        logResolvedMediaGenerationInput(options, currentVideoPath, result);
+        return result;
+      }
+      const result = await toResolvedMediaGenerationInput(
+        mpvClient,
+        streamOpenFilename,
+        kind,
+        'stream-open-filename',
+        false,
+      );
+      logResolvedMediaGenerationInput(options, currentVideoPath, result);
+      return result;
+    }
+    if (streamOpenFilename) {
+      const result = await toResolvedMediaGenerationInput(
+        mpvClient,
+        streamOpenFilename,
+        kind,
+        'stream-open-filename',
+        isRemoteMediaPath(streamOpenFilename),
+      );
+      logResolvedMediaGenerationInput(options, currentVideoPath, result);
+      return result;
+    }
+  } catch {
+    // Fall back to the current path when mpv does not expose a resolved stream URL.
+  }
+
+  const result = await toResolvedMediaGenerationInput(
+    mpvClient,
+    currentVideoPath,
+    kind,
+    'current-path',
+    false,
+  );
+  logResolvedMediaGenerationInput(options, currentVideoPath, result);
+  return result;
+}
+
 export async function resolveMediaGenerationInputPath(
   mpvClient: Pick<MpvClient, 'currentVideoPath' | 'requestProperty'> | null | undefined,
   kind: MediaGenerationKind = 'video',

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

@@ -409,3 +409,59 @@ test('NoteUpdateWorkflow uses subtitle sidebar context for sentence media timing
   assert.deepEqual(imageContext, sidebarContext);
   assert.equal(miscInfoStartTime, 10);
 });
+
+test('NoteUpdateWorkflow queues media updates when YouTube cache is pending', async () => {
+  const harness = createWorkflowHarness();
+  const queuedUpdates: Array<{
+    noteId: number;
+    noteInfo: NoteUpdateWorkflowNoteInfo;
+    context?: SubtitleMiningContext;
+    label: string | number;
+  }> = [];
+  const mediaCalls: string[] = [];
+
+  harness.deps.client.notesInfo = async () =>
+    [
+      {
+        noteId: 42,
+        fields: {
+          Expression: { value: 'taberu' },
+          Sentence: { value: '' },
+          SentenceAudio: { value: '' },
+          Picture: { value: '' },
+        },
+      },
+    ] satisfies NoteUpdateWorkflowNoteInfo[];
+  harness.deps.getConfig = () => ({
+    fields: {
+      sentence: 'Sentence',
+      image: 'Picture',
+    },
+    media: {
+      generateAudio: true,
+      generateImage: true,
+    },
+    behavior: {},
+  });
+  harness.deps.generateAudio = async () => {
+    mediaCalls.push('audio');
+    return Buffer.from('audio');
+  };
+  harness.deps.generateImage = async () => {
+    mediaCalls.push('image');
+    return Buffer.from('image');
+  };
+  harness.deps.queuePendingYoutubeMediaUpdate = async (job) => {
+    queuedUpdates.push(job);
+    return true;
+  };
+
+  await harness.workflow.execute(42);
+
+  assert.deepEqual(mediaCalls, []);
+  assert.equal(queuedUpdates.length, 1);
+  assert.equal(queuedUpdates[0]?.noteId, 42);
+  assert.equal(queuedUpdates[0]?.label, 'taberu');
+  assert.equal(queuedUpdates[0]?.context, undefined);
+  assert.deepEqual(harness.updates, [{ noteId: 42, fields: { Sentence: 'subtitle-text' } }]);
+});

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

@@ -85,6 +85,12 @@ export interface NoteUpdateWorkflowDeps {
   ) => Promise<Buffer | null>;
   formatMiscInfoPattern: (fallbackFilename: string, startTimeSeconds?: number) => string;
   consumeSubtitleMiningContext?: () => SubtitleMiningContext | null;
+  queuePendingYoutubeMediaUpdate?: (job: {
+    noteId: number;
+    noteInfo: NoteUpdateWorkflowNoteInfo;
+    context?: SubtitleMiningContext;
+    label: string | number;
+  }) => Promise<boolean>;
   addConfiguredTagsToNote: (noteId: number) => Promise<void>;
   showNotification: (noteId: number, label: string | number) => Promise<void>;
   showOsdNotification: (message: string) => void;
@@ -195,6 +201,7 @@ export class NoteUpdateWorkflow {
         sentenceField,
         config.fields?.sentence,
       );
+      const noteLabel = hasExpressionText ? expressionText : noteId;
 
       const currentSubtitleText = subtitleMiningContext?.text ?? this.deps.getCurrentSubtitleText();
       if (sentenceField && currentSubtitleText) {
@@ -227,7 +234,19 @@ export class NoteUpdateWorkflow {
         }
       }
 
-      if (config.media?.generateAudio) {
+      const generateAudio = config.media?.generateAudio !== false;
+      const generateImage = config.media?.generateImage !== false;
+      const mediaCacheQueued =
+        (generateAudio || generateImage) && this.deps.queuePendingYoutubeMediaUpdate
+          ? await this.deps.queuePendingYoutubeMediaUpdate({
+              noteId,
+              noteInfo,
+              context: subtitleMiningContext ?? undefined,
+              label: noteLabel,
+            })
+          : false;
+
+      if (!mediaCacheQueued && generateAudio) {
         try {
           const audioFilename = this.deps.generateAudioFilename();
           const audioBuffer = await this.deps.generateAudio(subtitleMiningContext ?? undefined);
@@ -252,7 +271,7 @@ export class NoteUpdateWorkflow {
         }
       }
 
-      if (config.media?.generateImage) {
+      if (!mediaCacheQueued && generateImage) {
         try {
           const animatedLeadInSeconds = await this.deps.getAnimatedImageLeadInSeconds(noteInfo);
           const imageFilename = this.deps.generateImageFilename();
@@ -287,7 +306,7 @@ export class NoteUpdateWorkflow {
         }
       }
 
-      if (config.fields?.miscInfo) {
+      if (!mediaCacheQueued && config.fields?.miscInfo) {
         const miscInfo = this.deps.formatMiscInfoPattern(
           miscInfoFilename || '',
           subtitleMiningContext?.startTime ?? this.deps.getCurrentSubtitleStart(),
@@ -305,8 +324,8 @@ export class NoteUpdateWorkflow {
       if (updatePerformed) {
         await this.deps.client.updateNoteFields(noteId, updatedFields);
         await this.deps.addConfiguredTagsToNote(noteId);
-        this.deps.logInfo('Updated card fields for:', hasExpressionText ? expressionText : noteId);
-        await this.deps.showNotification(noteId, hasExpressionText ? expressionText : noteId);
+        this.deps.logInfo('Updated card fields for:', noteLabel);
+        await this.deps.showNotification(noteId, noteLabel);
       }
 
       if (shouldRunFieldGrouping && hasExpressionText && duplicateNoteId !== null) {

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

@@ -0,0 +1,193 @@
+import assert from 'node:assert/strict';
+import test from 'node:test';
+
+import type { AnkiConnectConfig } from '../types/anki';
+import {
+  PendingYoutubeMediaQueue,
+  type PendingYoutubeMediaQueueDeps,
+} from './pending-youtube-media-queue';
+
+function createDeps(
+  overrides: Partial<PendingYoutubeMediaQueueDeps> = {},
+): PendingYoutubeMediaQueueDeps {
+  const warnings: unknown[][] = [];
+  const deps: PendingYoutubeMediaQueueDeps & { warnings: unknown[][] } = {
+    client: {
+      notesInfo: async () => [],
+      updateNoteFields: async () => {},
+      storeMediaFile: async () => {},
+    },
+    mediaGenerator: {
+      generateAudio: async () => Buffer.from('audio'),
+      generateScreenshot: async () => Buffer.from('image'),
+      generateAnimatedImage: async () => Buffer.from('image'),
+    },
+    getConfig: () =>
+      ({
+        media: { generateAudio: true, generateImage: true },
+        fields: {},
+      }) as AnkiConnectConfig,
+    getCurrentVideoPath: () => 'https://www.youtube.com/watch?v=abc123',
+    getCachedMediaPath: async () => null,
+    shouldRequireRemoteMediaCache: () => true,
+    getSubtitleMediaRange: () => ({ startTime: 1, endTime: 2 }),
+    getResolvedSentenceAudioFieldName: () => 'SentenceAudio',
+    resolveConfiguredFieldName: () => 'Picture',
+    mergeFieldValue: (_existing, newValue) => newValue,
+    getAnimatedImageLeadInSeconds: async () => 0,
+    generateAudioFilename: () => 'audio.mp3',
+    generateImageFilename: () => 'image.webp',
+    formatMiscInfoPatternForMediaPath: () => '',
+    showStatusNotification: () => {},
+    showNotification: async () => {},
+    logInfo: () => {},
+    logWarn: (...args) => {
+      warnings.push(args);
+    },
+    logError: () => {},
+    warnings,
+    ...overrides,
+  };
+  return deps;
+}
+
+test('PendingYoutubeMediaQueue treats cache lookup failures as an immediate generation fallback', async () => {
+  const deps = createDeps({
+    getCachedMediaPath: async () => {
+      throw new Error('cache unavailable');
+    },
+  });
+  const queue = new PendingYoutubeMediaQueue(deps);
+
+  const queued = await queue.queueFromNote({
+    noteId: 42,
+    noteInfo: { noteId: 42, fields: {} },
+    label: 'demo',
+  });
+
+  assert.equal(queued, false);
+  assert.deepEqual((deps as typeof deps & { warnings: unknown[][] }).warnings, [
+    [
+      'Failed to read YouTube cache state; falling back to immediate media generation:',
+      'cache unavailable',
+    ],
+  ]);
+});
+
+test('PendingYoutubeMediaQueue drains matching queued jobs when the cache download fails', async () => {
+  const statusMessages: string[] = [];
+  const notifications: Array<{ noteId: number; label: string | number; suffix?: string }> = [];
+  const updatedNotes: number[] = [];
+  const deps = createDeps({
+    client: {
+      notesInfo: async (noteIds) =>
+        noteIds.map((noteId) => ({
+          noteId,
+          fields: {
+            SentenceAudio: { value: '' },
+            Picture: { value: '' },
+          },
+        })),
+      updateNoteFields: async (noteId) => {
+        updatedNotes.push(noteId);
+      },
+      storeMediaFile: async () => {},
+    },
+    showStatusNotification: (message) => {
+      statusMessages.push(message);
+    },
+    showNotification: async (noteId, label, suffix) => {
+      notifications.push({ noteId, label, suffix });
+    },
+  });
+  const queue = new PendingYoutubeMediaQueue(deps);
+
+  queue.enqueue({
+    sourceUrl: 'https://www.youtube.com/watch?v=abc123',
+    noteId: 42,
+    startTime: 1,
+    endTime: 2,
+    label: 'queued',
+    generateAudio: true,
+    generateImage: true,
+  });
+
+  await queue.handleFailed('https://youtu.be/abc123');
+  await queue.handleReady('https://youtu.be/abc123', '/tmp/media.mkv');
+
+  assert.deepEqual(updatedNotes, []);
+  assert.deepEqual(notifications, [{ noteId: 42, label: 'queued', suffix: 'media cache failed' }]);
+  assert.equal(
+    statusMessages.includes('YouTube media cache failed. Media was not added to 1 queued card.'),
+    true,
+  );
+});
+
+test('PendingYoutubeMediaQueue defaults missing media flags to enabled when queuing from notes', async () => {
+  const updatedNotes: Array<{ noteId: number; fields: Record<string, string> }> = [];
+  const storedMedia: string[] = [];
+  const deps = createDeps({
+    client: {
+      notesInfo: async (noteIds) =>
+        noteIds.map((noteId) => ({
+          noteId,
+          fields: {
+            SentenceAudio: { value: '' },
+            Picture: { value: '' },
+          },
+        })),
+      updateNoteFields: async (noteId, fields) => {
+        updatedNotes.push({ noteId, fields });
+      },
+      storeMediaFile: async (filename) => {
+        storedMedia.push(filename);
+      },
+    },
+    getConfig: () => ({ media: {}, fields: { image: 'Picture' } }) as AnkiConnectConfig,
+  });
+  const queue = new PendingYoutubeMediaQueue(deps);
+
+  const queued = await queue.queueFromNote({
+    noteId: 42,
+    noteInfo: { noteId: 42, fields: {} },
+    label: 'demo',
+  });
+  await queue.handleReady('https://youtu.be/abc123', '/tmp/media.mkv');
+
+  assert.equal(queued, true);
+  assert.equal(updatedNotes.length, 1);
+  assert.equal(storedMedia.length, 2);
+  assert.match(updatedNotes[0]?.fields.SentenceAudio ?? '', /^\[sound:audio\.mp3\]$/);
+  assert.match(updatedNotes[0]?.fields.Picture ?? '', /^<img src="image\.webp">$/);
+});
+
+test('PendingYoutubeMediaQueue only announces a download once per source while jobs collect', () => {
+  const statusMessages: string[] = [];
+  const deps = createDeps({
+    showStatusNotification: (message) => {
+      statusMessages.push(message);
+    },
+  });
+  const queue = new PendingYoutubeMediaQueue(deps);
+
+  queue.enqueue({
+    sourceUrl: 'https://youtu.be/abc123',
+    noteId: 1,
+    startTime: 1,
+    endTime: 2,
+    label: 'first',
+    generateAudio: true,
+    generateImage: false,
+  });
+  queue.enqueue({
+    sourceUrl: 'https://www.youtube.com/watch?v=abc123',
+    noteId: 2,
+    startTime: 3,
+    endTime: 4,
+    label: 'second',
+    generateAudio: false,
+    generateImage: true,
+  });
+
+  assert.equal(statusMessages.length, 1);
+});

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

@@ -0,0 +1,391 @@
+import { DEFAULT_ANKI_CONNECT_CONFIG } from '../config';
+import type { MediaInput } from '../media-input';
+import type { MediaGenerator } from '../media-generator';
+import type { AnkiConnectConfig } from '../types/anki';
+import type { SubtitleMiningContext } from '../types/subtitle';
+import { youtubeMediaUrlsMatch, type PendingYoutubeMediaUpdate } from './pending-youtube-media';
+import type { MediaGenerationInputResolverOptions } from './media-source';
+
+type PendingYoutubeMediaUpdateResult = 'updated' | 'partial' | 'failed';
+
+export interface PendingYoutubeMediaQueueReadyOptions {
+  notifyNoQueued?: boolean;
+}
+
+export interface PendingYoutubeMediaQueueFailedOptions {
+  notifyStatus?: boolean;
+}
+
+export interface PendingYoutubeMediaNoteInfo {
+  noteId: number;
+  fields: Record<string, { value: string }>;
+}
+
+export interface PendingYoutubeMediaQueueDeps {
+  client: {
+    notesInfo(noteIds: number[]): Promise<unknown>;
+    updateNoteFields(noteId: number, fields: Record<string, string>): Promise<void>;
+    storeMediaFile(filename: string, data: Buffer): Promise<void>;
+  };
+  mediaGenerator: Pick<
+    MediaGenerator,
+    'generateAudio' | 'generateScreenshot' | 'generateAnimatedImage'
+  >;
+  getConfig: () => AnkiConnectConfig;
+  getCurrentVideoPath: () => string | undefined;
+  getCachedMediaPath: MediaGenerationInputResolverOptions['getCachedMediaPath'] | null;
+  shouldRequireRemoteMediaCache: () => boolean;
+  getSubtitleMediaRange: (context?: SubtitleMiningContext) => {
+    startTime: number;
+    endTime: number;
+  };
+  getResolvedSentenceAudioFieldName: (noteInfo: PendingYoutubeMediaNoteInfo) => string | null;
+  resolveConfiguredFieldName: (
+    noteInfo: PendingYoutubeMediaNoteInfo,
+    ...preferredNames: (string | undefined)[]
+  ) => string | null;
+  mergeFieldValue: (existing: string, newValue: string, overwrite: boolean) => string;
+  getAnimatedImageLeadInSeconds: (noteInfo: PendingYoutubeMediaNoteInfo) => Promise<number>;
+  generateAudioFilename: () => string;
+  generateImageFilename: () => string;
+  formatMiscInfoPatternForMediaPath: (
+    fallbackFilename: string,
+    startTimeSeconds: number | undefined,
+    mediaPath: string,
+    mediaTitle?: string,
+  ) => string;
+  showStatusNotification: (message: string) => void;
+  showNotification: (noteId: number, label: string | number, errorSuffix?: string) => Promise<void>;
+  logInfo: (message: string, ...args: unknown[]) => void;
+  logWarn: (message: string, ...args: unknown[]) => void;
+  logError: (message: string, ...args: unknown[]) => void;
+}
+
+function trimToNonEmptyString(value: unknown): string | null {
+  if (typeof value !== 'string') {
+    return null;
+  }
+  const trimmed = value.trim();
+  return trimmed.length > 0 ? trimmed : null;
+}
+
+function shouldGenerateAudio(config: AnkiConnectConfig): boolean {
+  return config.media?.generateAudio !== false;
+}
+
+function shouldGenerateImage(config: AnkiConnectConfig): boolean {
+  return config.media?.generateImage !== false;
+}
+
+export class PendingYoutubeMediaQueue {
+  private updates: PendingYoutubeMediaUpdate[] = [];
+
+  constructor(private readonly deps: PendingYoutubeMediaQueueDeps) {}
+
+  enqueue(job: PendingYoutubeMediaUpdate): void {
+    if (!job.generateAudio && !job.generateImage) {
+      return;
+    }
+    const isFirstQueuedForSource = !this.updates.some((existing) =>
+      youtubeMediaUrlsMatch(existing.sourceUrl, job.sourceUrl),
+    );
+    this.updates.push(job);
+    this.deps.logInfo('Queued YouTube media update for note:', job.noteId);
+    if (isFirstQueuedForSource) {
+      this.deps.showStatusNotification(
+        'YouTube media cache is still downloading. Card media will be added when the cache is ready.',
+      );
+    }
+  }
+
+  async queueFromNote(job: {
+    noteId: number;
+    noteInfo: PendingYoutubeMediaNoteInfo;
+    context?: SubtitleMiningContext;
+    label: string | number;
+  }): Promise<boolean> {
+    const sourceUrl = trimToNonEmptyString(this.deps.getCurrentVideoPath());
+    const getCachedMediaPath = this.deps.getCachedMediaPath;
+    if (!sourceUrl || this.deps.shouldRequireRemoteMediaCache() !== true || !getCachedMediaPath) {
+      return false;
+    }
+
+    let cachedPath: string | null = null;
+    try {
+      cachedPath = await getCachedMediaPath(sourceUrl, 'video');
+    } catch (error) {
+      this.deps.logWarn(
+        'Failed to read YouTube cache state; falling back to immediate media generation:',
+        error instanceof Error ? error.message : String(error),
+      );
+      return false;
+    }
+    if (cachedPath) {
+      return false;
+    }
+
+    const config = this.deps.getConfig();
+    const mediaRange = this.deps.getSubtitleMediaRange(job.context);
+    this.enqueue({
+      sourceUrl,
+      noteId: job.noteId,
+      startTime: mediaRange.startTime,
+      endTime: mediaRange.endTime,
+      label: job.label,
+      audioFieldName: this.deps.getResolvedSentenceAudioFieldName(job.noteInfo) ?? undefined,
+      imageFieldName:
+        this.deps.resolveConfiguredFieldName(
+          job.noteInfo,
+          config.fields?.image,
+          DEFAULT_ANKI_CONNECT_CONFIG.fields.image,
+        ) ?? undefined,
+      miscInfoFieldName:
+        this.deps.resolveConfiguredFieldName(job.noteInfo, config.fields?.miscInfo) ?? undefined,
+      generateAudio: shouldGenerateAudio(config),
+      generateImage: shouldGenerateImage(config),
+    });
+    return true;
+  }
+
+  async handleReady(
+    sourceUrl: string,
+    cachedPath: string,
+    options: PendingYoutubeMediaQueueReadyOptions = {},
+  ): Promise<void> {
+    const jobs = this.takeMatchingUpdates(sourceUrl);
+    if (jobs.length === 0) {
+      if (options.notifyNoQueued !== false) {
+        this.deps.showStatusNotification('YouTube media cache ready.');
+      }
+      return;
+    }
+
+    this.deps.showStatusNotification(
+      `YouTube media cache ready. Adding media to ${jobs.length} queued card${
+        jobs.length === 1 ? '' : 's'
+      }.`,
+    );
+
+    let updatedCount = 0;
+    let partialCount = 0;
+    let failedCount = 0;
+    for (const job of jobs) {
+      try {
+        const result = await this.applyUpdate(job, cachedPath);
+        if (result === 'updated') {
+          updatedCount += 1;
+        } else if (result === 'partial') {
+          partialCount += 1;
+        } else {
+          failedCount += 1;
+        }
+      } catch (error) {
+        failedCount += 1;
+        this.deps.logError(
+          'Failed to apply queued YouTube media update:',
+          error instanceof Error ? error.message : String(error),
+        );
+      }
+    }
+
+    if (partialCount > 0 || failedCount > 0) {
+      this.deps.showStatusNotification(
+        `Queued YouTube media finished with ${updatedCount} updated, ${partialCount} partial, and ${failedCount} failed.`,
+      );
+    }
+  }
+
+  async handleFailed(
+    sourceUrl: string,
+    options: PendingYoutubeMediaQueueFailedOptions = {},
+  ): Promise<void> {
+    const jobs = this.takeMatchingUpdates(sourceUrl);
+    if (jobs.length === 0) {
+      if (options.notifyStatus !== false) {
+        this.deps.showStatusNotification('YouTube media cache failed.');
+      }
+      return;
+    }
+
+    if (options.notifyStatus !== false) {
+      this.deps.showStatusNotification(
+        `YouTube media cache failed. Media was not added to ${jobs.length} queued card${
+          jobs.length === 1 ? '' : 's'
+        }.`,
+      );
+    }
+    this.deps.logWarn('Discarding queued YouTube media updates after cache failure:', jobs.length);
+
+    for (const job of jobs) {
+      try {
+        await this.deps.showNotification(job.noteId, job.label, 'media cache failed');
+      } catch (error) {
+        this.deps.logWarn(
+          'Failed to show queued YouTube media failure notification:',
+          error instanceof Error ? error.message : String(error),
+        );
+      }
+    }
+  }
+
+  private takeMatchingUpdates(sourceUrl: string): PendingYoutubeMediaUpdate[] {
+    const matched: PendingYoutubeMediaUpdate[] = [];
+    const remaining: PendingYoutubeMediaUpdate[] = [];
+    for (const job of this.updates) {
+      if (youtubeMediaUrlsMatch(job.sourceUrl, sourceUrl)) {
+        matched.push(job);
+      } else {
+        remaining.push(job);
+      }
+    }
+    this.updates = remaining;
+    return matched;
+  }
+
+  private async applyUpdate(
+    job: PendingYoutubeMediaUpdate,
+    cachedPath: string,
+  ): Promise<PendingYoutubeMediaUpdateResult> {
+    const notesInfoResult = await this.deps.client.notesInfo([job.noteId]);
+    const notesInfo = notesInfoResult as unknown as PendingYoutubeMediaNoteInfo[];
+    const noteInfo = notesInfo[0];
+    if (!noteInfo) {
+      this.deps.logWarn('Queued YouTube media target note not found:', job.noteId);
+      return 'failed';
+    }
+
+    const config = this.deps.getConfig();
+    const mediaFields: Record<string, string> = {};
+    const errors: string[] = [];
+    let miscInfoFilename: string | null = null;
+    const cachedMediaInput: MediaInput = {
+      path: cachedPath,
+      source: 'youtube-cache',
+    };
+
+    if (job.generateAudio) {
+      try {
+        const audioFilename = this.deps.generateAudioFilename();
+        const audioBuffer = await this.deps.mediaGenerator.generateAudio(
+          cachedMediaInput,
+          job.startTime,
+          job.endTime,
+          config.media?.audioPadding,
+          undefined,
+        );
+        if (audioBuffer) {
+          await this.deps.client.storeMediaFile(audioFilename, audioBuffer);
+          const audioField =
+            job.audioFieldName || this.deps.getResolvedSentenceAudioFieldName(noteInfo) || null;
+          if (audioField) {
+            const existingAudio = noteInfo.fields[audioField]?.value || '';
+            mediaFields[audioField] = this.deps.mergeFieldValue(
+              existingAudio,
+              `[sound:${audioFilename}]`,
+              config.behavior?.overwriteAudio !== false,
+            );
+          }
+          miscInfoFilename = audioFilename;
+        }
+      } catch (error) {
+        errors.push('audio');
+        this.deps.logError('Failed to generate queued YouTube audio:', (error as Error).message);
+      }
+    }
+
+    if (job.generateImage) {
+      try {
+        const imageFilename = this.deps.generateImageFilename();
+        const animatedLeadInSeconds = await this.deps.getAnimatedImageLeadInSeconds(noteInfo);
+        const imageBuffer = await this.generateImageFromInput(
+          cachedMediaInput,
+          job.startTime,
+          job.endTime,
+          animatedLeadInSeconds,
+        );
+        if (imageBuffer) {
+          await this.deps.client.storeMediaFile(imageFilename, imageBuffer);
+          const imageField =
+            job.imageFieldName ||
+            this.deps.resolveConfiguredFieldName(
+              noteInfo,
+              config.fields?.image,
+              DEFAULT_ANKI_CONNECT_CONFIG.fields.image,
+            );
+          if (imageField) {
+            const existingImage = noteInfo.fields[imageField]?.value || '';
+            mediaFields[imageField] = this.deps.mergeFieldValue(
+              existingImage,
+              `<img src="${imageFilename}">`,
+              config.behavior?.overwriteImage !== false,
+            );
+          } else {
+            this.deps.logWarn(
+              'Image field not found on queued YouTube media note, skipping image update',
+            );
+          }
+          miscInfoFilename = imageFilename;
+        }
+      } catch (error) {
+        errors.push('image');
+        this.deps.logError('Failed to generate queued YouTube image:', (error as Error).message);
+      }
+    }
+
+    if (config.fields?.miscInfo && miscInfoFilename) {
+      const miscInfoField =
+        job.miscInfoFieldName ||
+        this.deps.resolveConfiguredFieldName(noteInfo, config.fields.miscInfo);
+      const miscInfo = this.deps.formatMiscInfoPatternForMediaPath(
+        miscInfoFilename,
+        job.startTime,
+        job.sourceUrl,
+      );
+      if (miscInfoField && miscInfo) {
+        mediaFields[miscInfoField] = miscInfo;
+      }
+    }
+
+    if (Object.keys(mediaFields).length === 0) {
+      return 'failed';
+    }
+
+    await this.deps.client.updateNoteFields(job.noteId, mediaFields);
+    const errorSuffix = errors.length > 0 ? `${errors.join(', ')} failed` : undefined;
+    await this.deps.showNotification(job.noteId, job.label, errorSuffix);
+    this.deps.logInfo('Applied queued YouTube media update for note:', job.noteId);
+    return errors.length === 0 ? 'updated' : 'partial';
+  }
+
+  private async generateImageFromInput(
+    videoPath: MediaInput,
+    startTime: number,
+    endTime: number,
+    animatedLeadInSeconds = 0,
+  ): Promise<Buffer | null> {
+    const config = this.deps.getConfig();
+    if (config.media?.imageType === 'avif') {
+      return this.deps.mediaGenerator.generateAnimatedImage(
+        videoPath,
+        startTime,
+        endTime,
+        config.media?.audioPadding,
+        {
+          fps: config.media?.animatedFps,
+          maxWidth: config.media?.animatedMaxWidth,
+          maxHeight: config.media?.animatedMaxHeight,
+          crf: config.media?.animatedCrf,
+          leadingStillDuration: animatedLeadInSeconds,
+        },
+      );
+    }
+
+    const timestamp = startTime + (endTime - startTime) / 2;
+    return this.deps.mediaGenerator.generateScreenshot(videoPath, timestamp, {
+      format: config.media?.imageFormat as 'jpg' | 'png' | 'webp',
+      quality: config.media?.imageQuality,
+      maxWidth: config.media?.imageMaxWidth,
+      maxHeight: config.media?.imageMaxHeight,
+    });
+  }
+}

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

@@ -0,0 +1,53 @@
+export interface PendingYoutubeMediaUpdate {
+  sourceUrl: string;
+  noteId: number;
+  startTime: number;
+  endTime: number;
+  label: string | number;
+  audioFieldName?: string;
+  imageFieldName?: string;
+  miscInfoFieldName?: string;
+  generateAudio: boolean;
+  generateImage: boolean;
+}
+
+function trimToNonEmptyString(value: unknown): string | null {
+  if (typeof value !== 'string') {
+    return null;
+  }
+  const trimmed = value.trim();
+  return trimmed.length > 0 ? trimmed : null;
+}
+
+function getYoutubeVideoId(rawUrl: string): string | null {
+  try {
+    const parsed = new URL(rawUrl);
+    const host = parsed.hostname.toLowerCase();
+    if (host === 'youtu.be' || host.endsWith('.youtu.be')) {
+      return trimToNonEmptyString(parsed.pathname.replace(/^\/+/, '').split('/')[0]);
+    }
+    if (host === 'youtube.com' || host.endsWith('.youtube.com')) {
+      const watchId = trimToNonEmptyString(parsed.searchParams.get('v'));
+      if (watchId) {
+        return watchId;
+      }
+      const parts = parsed.pathname.split('/').filter(Boolean);
+      if (parts[0] === 'shorts' || parts[0] === 'embed' || parts[0] === 'live') {
+        return trimToNonEmptyString(parts[1]);
+      }
+    }
+  } catch {
+    return null;
+  }
+  return null;
+}
+
+export function youtubeMediaUrlsMatch(a: string, b: string): boolean {
+  if (a.trim() === b.trim()) {
+    return true;
+  }
+
+  const aVideoId = getYoutubeVideoId(a);
+  const bVideoId = getYoutubeVideoId(b);
+  return Boolean(aVideoId && bVideoId && aVideoId === bVideoId);
+}

src/config/config.test.ts

@@ -80,6 +80,8 @@ test('loads defaults when config is missing', () => {
   assert.equal('remoteControlDeviceName' in config.jellyfin, false);
   assert.equal('deviceId' in config.jellyfin, false);
   assert.equal('clientVersion' in config.jellyfin, false);
+  assert.equal(config.youtube.mediaCache.mode, 'direct');
+  assert.equal(config.youtube.mediaCache.maxHeight, 720);
   assert.equal(config.ai.enabled, false);
   assert.equal(config.ai.apiKeyCommand, '');
   assert.equal(config.texthooker.openBrowser, false);
@@ -1750,6 +1752,56 @@ test('parses global shortcuts and startup settings', () => {
   assert.equal(config.youtubeSubgen.fixWithAi, true);
 });
 
+test('parses YouTube media cache config and warns on invalid values', () => {
+  const validDir = makeTempDir();
+  fs.writeFileSync(
+    path.join(validDir, 'config.jsonc'),
+    `{
+      "youtube": {
+        "mediaCache": {
+          "mode": "background",
+          "maxHeight": 480
+        }
+      }
+    }`,
+    'utf-8',
+  );
+
+  const validService = new ConfigService(validDir);
+  assert.equal(validService.getConfig().youtube.mediaCache.mode, 'background');
+  assert.equal(validService.getConfig().youtube.mediaCache.maxHeight, 480);
+
+  const invalidDir = makeTempDir();
+  fs.writeFileSync(
+    path.join(invalidDir, 'config.jsonc'),
+    `{
+      "youtube": {
+        "mediaCache": {
+          "mode": "always",
+          "maxHeight": -1
+        }
+      }
+    }`,
+    'utf-8',
+  );
+
+  const invalidService = new ConfigService(invalidDir);
+  assert.equal(
+    invalidService.getConfig().youtube.mediaCache.mode,
+    DEFAULT_CONFIG.youtube.mediaCache.mode,
+  );
+  assert.equal(
+    invalidService.getConfig().youtube.mediaCache.maxHeight,
+    DEFAULT_CONFIG.youtube.mediaCache.maxHeight,
+  );
+  assert.ok(
+    invalidService.getWarnings().some((warning) => warning.path === 'youtube.mediaCache.mode'),
+  );
+  assert.ok(
+    invalidService.getWarnings().some((warning) => warning.path === 'youtube.mediaCache.maxHeight'),
+  );
+});
+
 test('parses controller settings with logical bindings and tuning knobs', () => {
   const dir = makeTempDir();
   fs.writeFileSync(

src/config/definitions/defaults-core.ts

@@ -111,6 +111,10 @@ export const CORE_DEFAULT_CONFIG: Pick<
   },
   youtube: {
     primarySubLanguages: ['ja', 'jpn'],
+    mediaCache: {
+      mode: 'direct',
+      maxHeight: 720,
+    },
   },
   subsync: {
     alass_path: '',

src/config/definitions/options-core.ts

@@ -119,6 +119,24 @@ export function buildCoreConfigOptionRegistry(
       description:
         'Comma-separated primary subtitle language priority for managed subtitle auto-selection.',
     },
+    {
+      path: 'youtube.mediaCache.mode',
+      kind: 'enum',
+      enumValues: ['direct', 'background'],
+      enumLabels: {
+        direct: 'Direct stream extraction',
+        background: 'Background media cache',
+      },
+      defaultValue: defaultConfig.youtube.mediaCache.mode,
+      description: 'How YouTube card audio/images are extracted.',
+    },
+    {
+      path: 'youtube.mediaCache.maxHeight',
+      kind: 'number',
+      defaultValue: defaultConfig.youtube.mediaCache.maxHeight,
+      description:
+        'Maximum video height downloaded for the YouTube background media cache. Set to 0 for unlimited.',
+    },
     {
       path: 'controller.enabled',
       kind: 'boolean',

src/config/definitions/shared.ts

@@ -32,6 +32,7 @@ export interface ConfigOptionRegistryEntry {
    * `osd` and `osd-system` in NOTIFICATION_TYPE_VALUES.
    */
   enumValues?: readonly string[];
+  enumLabels?: Record<string, string>;
   /**
    * Optional settings UI subset when legacy/runtime-valid enum options should remain
    * editable in config files but hidden from new UI choices, for example

src/config/resolve/core-domains.ts

@@ -297,6 +297,39 @@ export function applyCoreDomainConfig(context: ResolveContext): void {
         'Expected string array.',
       );
     }
+
+    if (isObject(src.youtube.mediaCache)) {
+      const mode = src.youtube.mediaCache.mode;
+      if (mode === 'direct' || mode === 'background') {
+        resolved.youtube.mediaCache.mode = mode;
+      } else if (mode !== undefined) {
+        warn(
+          'youtube.mediaCache.mode',
+          mode,
+          resolved.youtube.mediaCache.mode,
+          "Expected 'direct' or 'background'.",
+        );
+      }
+
+      const maxHeight = asNumber(src.youtube.mediaCache.maxHeight);
+      if (maxHeight !== undefined && Number.isInteger(maxHeight) && maxHeight >= 0) {
+        resolved.youtube.mediaCache.maxHeight = maxHeight;
+      } else if (src.youtube.mediaCache.maxHeight !== undefined) {
+        warn(
+          'youtube.mediaCache.maxHeight',
+          src.youtube.mediaCache.maxHeight,
+          resolved.youtube.mediaCache.maxHeight,
+          'Expected a whole number at least 0.',
+        );
+      }
+    } else if (src.youtube.mediaCache !== undefined) {
+      warn(
+        'youtube.mediaCache',
+        src.youtube.mediaCache,
+        resolved.youtube.mediaCache,
+        'Expected object.',
+      );
+    }
   }
 
   if (isObject(src.subsync)) {

src/config/settings/registry.test.ts

@@ -165,6 +165,20 @@ test('settings registry exposes specialized controls for config-assisted inputs'
   assert.equal(field('discordPresence.presenceStyle').control, 'select');
 });
 
+test('settings registry exposes YouTube media cache mode as a labeled select', () => {
+  const mediaCacheMode = field('youtube.mediaCache.mode');
+  const mediaCacheMaxHeight = field('youtube.mediaCache.maxHeight');
+
+  assert.equal(mediaCacheMode.control, 'select');
+  assert.deepEqual(mediaCacheMode.enumValues, ['direct', 'background']);
+  assert.deepEqual(mediaCacheMode.enumLabels, {
+    direct: 'Direct stream extraction',
+    background: 'Background media cache',
+  });
+  assert.equal(mediaCacheMaxHeight.control, 'number');
+  assert.equal(mediaCacheMaxHeight.defaultValue, 720);
+});
+
 test('settings registry exposes css declaration editor for primary and secondary subtitle appearance', () => {
   const primaryVisible = fields
     .filter(

src/config/settings/registry.ts

@@ -720,6 +720,7 @@ function fieldForLeaf(leaf: Leaf): ConfigSettingsField {
     ...(option?.settingsEnumValues || option?.enumValues
       ? { enumValues: option.settingsEnumValues ?? option.enumValues }
       : {}),
+    ...(option?.enumLabels ? { enumLabels: option.enumLabels } : {}),
     restartBehavior: restartBehaviorForPath(leaf.path),
     advanced:
       leaf.path.startsWith('controller.') ||

src/core/services/anki-jimaku.ts

@@ -40,6 +40,11 @@ export interface AnkiJimakuIpcRuntimeOptions {
   getAnkiIntegration: () => AnkiIntegration | null;
   setAnkiIntegration: (integration: AnkiIntegration | null) => void;
   getKnownWordCacheStatePath: () => string;
+  getCachedMediaPath?: (
+    currentVideoPath: string,
+    kind: 'audio' | 'video',
+  ) => Promise<string | null>;
+  shouldRequireRemoteMediaCache?: () => boolean;
   showDesktopNotification: (title: string, options: { body?: string; icon?: string }) => void;
   showOverlayNotification?: (payload: OverlayNotificationPayload) => void;
   createFieldGroupingCallback: () => (
@@ -107,6 +112,8 @@ export function registerAnkiJimakuIpcRuntime(
           mergeAiConfig(config.ai, config.ankiConnect?.ai) as AiConfig,
           undefined,
           options.showOverlayNotification,
+          options.getCachedMediaPath,
+          options.shouldRequireRemoteMediaCache,
         );
         integration.start();
         options.setAnkiIntegration(integration);

src/core/services/overlay-runtime-init.ts

@@ -25,6 +25,11 @@ type CreateAnkiIntegrationArgs = {
     data: KikuFieldGroupingRequestData,
   ) => Promise<KikuFieldGroupingChoice>;
   knownWordCacheStatePath: string;
+  getCachedMediaPath?: (
+    currentVideoPath: string,
+    kind: 'audio' | 'video',
+  ) => Promise<string | null>;
+  shouldRequireRemoteMediaCache?: () => boolean;
 };
 
 export type OverlayWindowTrackerOptions = {
@@ -65,6 +70,8 @@ function createDefaultAnkiIntegration(args: CreateAnkiIntegrationArgs): AnkiInte
     args.aiConfig,
     undefined,
     args.showOverlayNotification,
+    args.getCachedMediaPath,
+    args.shouldRequireRemoteMediaCache,
   );
 }
 
@@ -132,6 +139,11 @@ export function initializeOverlayRuntime(
       data: KikuFieldGroupingRequestData,
     ) => Promise<KikuFieldGroupingChoice>;
     getKnownWordCacheStatePath: () => string;
+    getCachedMediaPath?: (
+      currentVideoPath: string,
+      kind: 'audio' | 'video',
+    ) => Promise<string | null>;
+    shouldRequireRemoteMediaCache?: () => boolean;
     shouldStartAnkiIntegration?: () => boolean;
     createAnkiIntegration?: (args: CreateAnkiIntegrationArgs) => AnkiIntegrationLike;
     backendOverride: string | null;
@@ -166,6 +178,11 @@ export function initializeOverlayAnkiIntegration(options: {
     data: KikuFieldGroupingRequestData,
   ) => Promise<KikuFieldGroupingChoice>;
   getKnownWordCacheStatePath: () => string;
+  getCachedMediaPath?: (
+    currentVideoPath: string,
+    kind: 'audio' | 'video',
+  ) => Promise<string | null>;
+  shouldRequireRemoteMediaCache?: () => boolean;
   shouldStartAnkiIntegration?: () => boolean;
   createAnkiIntegration?: (args: CreateAnkiIntegrationArgs) => AnkiIntegrationLike;
 }): boolean {
@@ -200,6 +217,10 @@ export function initializeOverlayAnkiIntegration(options: {
     showOverlayNotification: options.showOverlayNotification,
     createFieldGroupingCallback: options.createFieldGroupingCallback,
     knownWordCacheStatePath: options.getKnownWordCacheStatePath(),
+    ...(options.getCachedMediaPath ? { getCachedMediaPath: options.getCachedMediaPath } : {}),
+    ...(options.shouldRequireRemoteMediaCache
+      ? { shouldRequireRemoteMediaCache: options.shouldRequireRemoteMediaCache }
+      : {}),
   });
   if (options.shouldStartAnkiIntegration?.() !== false) {
     integration.start();

src/core/services/youtube/media-cache.test.ts

@@ -0,0 +1,540 @@
+import assert from 'node:assert/strict';
+import { EventEmitter } from 'node:events';
+import * as fs from 'node:fs';
+import * as os from 'node:os';
+import * as path from 'node:path';
+import test from 'node:test';
+
+import { createYoutubeMediaCacheService } from './media-cache';
+
+class FakeYtDlpProcess extends EventEmitter {
+  killed = false;
+  stdout = new EventEmitter();
+  stderr = new EventEmitter();
+
+  kill(): boolean {
+    this.killed = true;
+    return true;
+  }
+}
+
+type SpawnCall = {
+  command: string;
+  args: string[];
+  options?: { stdio?: Array<'ignore' | 'pipe'> };
+};
+
+function makeTempCacheRoot(): string {
+  return fs.mkdtempSync(path.join(os.tmpdir(), 'subminer-youtube-media-cache-test-'));
+}
+
+test('YouTube media cache does nothing in direct mode', async () => {
+  const cacheRoot = makeTempCacheRoot();
+  const spawnCalls: SpawnCall[] = [];
+
+  try {
+    const cache = createYoutubeMediaCacheService({
+      cacheRoot,
+      getYtDlpCommand: () => 'yt-dlp',
+      spawn: (command, args) => {
+        spawnCalls.push({ command, args });
+        return new FakeYtDlpProcess();
+      },
+    });
+
+    cache.start('https://youtu.be/demo', { mode: 'direct' });
+
+    assert.deepEqual(spawnCalls, []);
+    assert.equal(await cache.getCachedMediaPath('https://youtu.be/demo'), null);
+  } finally {
+    fs.rmSync(cacheRoot, { recursive: true, force: true });
+  }
+});
+
+test('YouTube media cache exposes the downloaded file after the background job completes', async () => {
+  const cacheRoot = makeTempCacheRoot();
+  const spawnedProcesses: FakeYtDlpProcess[] = [];
+  const spawnCalls: SpawnCall[] = [];
+  const readyEvents: Array<{ url: string; path: string }> = [];
+  const startedEvents: Array<{ url: string }> = [];
+
+  try {
+    const cache = createYoutubeMediaCacheService({
+      cacheRoot,
+      getYtDlpCommand: () => 'yt-dlp',
+      onDownloadStarted: (event) => {
+        startedEvents.push(event);
+      },
+      onReady: (event) => {
+        readyEvents.push(event);
+      },
+      spawn: (command, args, options) => {
+        spawnCalls.push({ command, args, options });
+        const proc = new FakeYtDlpProcess();
+        spawnedProcesses.push(proc);
+        return proc;
+      },
+    });
+
+    cache.start('https://youtu.be/demo', { mode: 'background' });
+
+    assert.deepEqual(startedEvents, [{ url: 'https://youtu.be/demo' }]);
+    assert.equal(spawnCalls.length, 1);
+    assert.equal(spawnCalls[0]?.command, 'yt-dlp');
+    assert.ok(spawnCalls[0]?.args.includes('--no-playlist'));
+    assert.ok(spawnCalls[0]?.args.includes('--force-ipv4'));
+    assert.equal(spawnCalls[0]?.args[spawnCalls[0].args.indexOf('--retries') + 1], '5');
+    assert.equal(spawnCalls[0]?.args[spawnCalls[0].args.indexOf('--fragment-retries') + 1], '5');
+    assert.equal(spawnCalls[0]?.args[spawnCalls[0].args.indexOf('--extractor-retries') + 1], '5');
+    assert.ok(spawnCalls[0]?.args.includes('--merge-output-format'));
+    assert.equal(
+      spawnCalls[0]?.args[spawnCalls[0].args.indexOf('-f') + 1],
+      'bestvideo*[height<=720]+bestaudio/best[height<=720]',
+    );
+    assert.deepEqual(spawnCalls[0]?.options?.stdio, ['ignore', 'ignore', 'ignore']);
+    assert.equal(await cache.getCachedMediaPath('https://youtu.be/demo'), null);
+
+    const outputTemplate = spawnCalls[0]?.args[spawnCalls[0].args.indexOf('-o') + 1];
+    assert.equal(typeof outputTemplate, 'string');
+    const outputDir = path.dirname(outputTemplate!);
+    const outputPath = path.join(outputDir, 'media.mkv');
+    fs.mkdirSync(outputDir, { recursive: true });
+    fs.writeFileSync(outputPath, 'cached media');
+    spawnedProcesses[0]?.emit('close', 0);
+
+    await new Promise((resolve) => setImmediate(resolve));
+
+    assert.equal(await cache.getCachedMediaPath('https://youtu.be/demo'), outputPath);
+    assert.deepEqual(readyEvents, [{ url: 'https://youtu.be/demo', path: outputPath }]);
+  } finally {
+    fs.rmSync(cacheRoot, { recursive: true, force: true });
+  }
+});
+
+test('YouTube media cache clears the active cache when direct mode starts', async () => {
+  const cacheRoot = makeTempCacheRoot();
+  const spawnedProcesses: FakeYtDlpProcess[] = [];
+  const spawnCalls: SpawnCall[] = [];
+
+  try {
+    const cache = createYoutubeMediaCacheService({
+      cacheRoot,
+      getYtDlpCommand: () => 'yt-dlp',
+      spawn: (command, args, options) => {
+        spawnCalls.push({ command, args, options });
+        const proc = new FakeYtDlpProcess();
+        spawnedProcesses.push(proc);
+        return proc;
+      },
+    });
+
+    cache.start('https://youtu.be/background', { mode: 'background' });
+    const outputTemplate = spawnCalls[0]?.args[spawnCalls[0].args.indexOf('-o') + 1];
+    assert.equal(typeof outputTemplate, 'string');
+    const outputDir = path.dirname(outputTemplate!);
+    const outputPath = path.join(outputDir, 'media.mkv');
+    fs.mkdirSync(outputDir, { recursive: true });
+    fs.writeFileSync(outputPath, 'cached media');
+    spawnedProcesses[0]?.emit('close', 0);
+
+    await new Promise((resolve) => setImmediate(resolve));
+    assert.equal(await cache.getActiveCachedMediaPath(), outputPath);
+
+    cache.start('https://youtu.be/direct', { mode: 'direct' });
+
+    assert.equal(await cache.getActiveCachedMediaPath(), null);
+    assert.equal(await cache.getCachedMediaPath('https://youtu.be/background'), outputPath);
+  } finally {
+    fs.rmSync(cacheRoot, { recursive: true, force: true });
+  }
+});
+
+test('YouTube media cache cancels the active download when direct mode starts', async () => {
+  const cacheRoot = makeTempCacheRoot();
+  const spawnedProcesses: FakeYtDlpProcess[] = [];
+  const spawnCalls: SpawnCall[] = [];
+  const readyEvents: Array<{ url: string; path: string }> = [];
+
+  try {
+    const cache = createYoutubeMediaCacheService({
+      cacheRoot,
+      getYtDlpCommand: () => 'yt-dlp',
+      onReady: (event) => {
+        readyEvents.push(event);
+      },
+      spawn: (command, args, options) => {
+        spawnCalls.push({ command, args, options });
+        const proc = new FakeYtDlpProcess();
+        spawnedProcesses.push(proc);
+        return proc;
+      },
+    });
+
+    cache.start('https://youtu.be/background', { mode: 'background' });
+    const outputTemplate = spawnCalls[0]?.args[spawnCalls[0].args.indexOf('-o') + 1];
+    assert.equal(typeof outputTemplate, 'string');
+    const outputDir = path.dirname(outputTemplate!);
+    const outputPath = path.join(outputDir, 'media.mkv');
+    fs.mkdirSync(outputDir, { recursive: true });
+    fs.writeFileSync(outputPath, 'cached media');
+
+    cache.start('https://youtu.be/direct', { mode: 'direct' });
+    spawnedProcesses[0]?.emit('close', 0);
+
+    await new Promise((resolve) => setImmediate(resolve));
+
+    assert.equal(spawnedProcesses[0]?.killed, true);
+    assert.deepEqual(readyEvents, []);
+    assert.equal(await cache.getActiveCachedMediaPath(), null);
+    assert.equal(await cache.getCachedMediaPath('https://youtu.be/background'), null);
+  } finally {
+    fs.rmSync(cacheRoot, { recursive: true, force: true });
+  }
+});
+
+test('YouTube media cache reports failed background downloads', async () => {
+  const cacheRoot = makeTempCacheRoot();
+  const spawnedProcesses: FakeYtDlpProcess[] = [];
+  const failedEvents: Array<{ url: string }> = [];
+
+  try {
+    const cache = createYoutubeMediaCacheService({
+      cacheRoot,
+      getYtDlpCommand: () => 'yt-dlp',
+      onFailed: (event) => {
+        failedEvents.push(event);
+      },
+      spawn: () => {
+        const proc = new FakeYtDlpProcess();
+        spawnedProcesses.push(proc);
+        return proc;
+      },
+    });
+
+    cache.start('https://youtu.be/demo', { mode: 'background' });
+    spawnedProcesses[0]?.emit('close', 1);
+
+    await new Promise((resolve) => setImmediate(resolve));
+
+    assert.deepEqual(failedEvents, [{ url: 'https://youtu.be/demo' }]);
+    assert.equal(await cache.getCachedMediaPath('https://youtu.be/demo'), null);
+  } finally {
+    fs.rmSync(cacheRoot, { recursive: true, force: true });
+  }
+});
+
+test('YouTube media cache only reports a failed download once when error is followed by close', async () => {
+  const cacheRoot = makeTempCacheRoot();
+  const spawnedProcesses: FakeYtDlpProcess[] = [];
+  const failedEvents: Array<{ url: string }> = [];
+
+  try {
+    const cache = createYoutubeMediaCacheService({
+      cacheRoot,
+      getYtDlpCommand: () => 'yt-dlp',
+      onFailed: (event) => {
+        failedEvents.push(event);
+      },
+      spawn: () => {
+        const proc = new FakeYtDlpProcess();
+        spawnedProcesses.push(proc);
+        return proc;
+      },
+    });
+
+    cache.start('https://youtu.be/demo', { mode: 'background' });
+    spawnedProcesses[0]?.emit('error', new Error('spawn failed'));
+    spawnedProcesses[0]?.emit('close', 1);
+
+    await new Promise((resolve) => setImmediate(resolve));
+
+    assert.deepEqual(failedEvents, [{ url: 'https://youtu.be/demo' }]);
+  } finally {
+    fs.rmSync(cacheRoot, { recursive: true, force: true });
+  }
+});
+
+test('YouTube media cache can disable the download height cap', () => {
+  const cacheRoot = makeTempCacheRoot();
+  const spawnCalls: SpawnCall[] = [];
+
+  try {
+    const cache = createYoutubeMediaCacheService({
+      cacheRoot,
+      getYtDlpCommand: () => 'yt-dlp',
+      spawn: (command, args, options) => {
+        spawnCalls.push({ command, args, options });
+        return new FakeYtDlpProcess();
+      },
+    });
+
+    cache.start('https://youtu.be/demo', { mode: 'background', maxHeight: 0 });
+
+    assert.equal(spawnCalls.length, 1);
+    assert.equal(
+      spawnCalls[0]?.args[spawnCalls[0].args.indexOf('-f') + 1],
+      'bestvideo*+bestaudio/best',
+    );
+  } finally {
+    fs.rmSync(cacheRoot, { recursive: true, force: true });
+  }
+});
+
+test('YouTube media cache applies the configured download height cap', () => {
+  const cacheRoot = makeTempCacheRoot();
+  const spawnCalls: SpawnCall[] = [];
+
+  try {
+    const cache = createYoutubeMediaCacheService({
+      cacheRoot,
+      getYtDlpCommand: () => 'yt-dlp',
+      spawn: (command, args, options) => {
+        spawnCalls.push({ command, args, options });
+        return new FakeYtDlpProcess();
+      },
+    });
+
+    cache.start('https://youtu.be/demo', { mode: 'background', maxHeight: 480 });
+
+    assert.equal(spawnCalls.length, 1);
+    assert.equal(
+      spawnCalls[0]?.args[spawnCalls[0].args.indexOf('-f') + 1],
+      'bestvideo*[height<=480]+bestaudio/best[height<=480]',
+    );
+  } finally {
+    fs.rmSync(cacheRoot, { recursive: true, force: true });
+  }
+});
+
+test('YouTube media cache restarts ready sessions when the height cap changes', async () => {
+  const cacheRoot = makeTempCacheRoot();
+  const spawnedProcesses: FakeYtDlpProcess[] = [];
+  const spawnCalls: SpawnCall[] = [];
+
+  try {
+    const cache = createYoutubeMediaCacheService({
+      cacheRoot,
+      getYtDlpCommand: () => 'yt-dlp',
+      spawn: (command, args, options) => {
+        spawnCalls.push({ command, args, options });
+        const proc = new FakeYtDlpProcess();
+        spawnedProcesses.push(proc);
+        return proc;
+      },
+    });
+
+    cache.start('https://youtu.be/demo', { mode: 'background', maxHeight: 720 });
+    const firstOutputTemplate = spawnCalls[0]?.args[spawnCalls[0].args.indexOf('-o') + 1];
+    assert.equal(typeof firstOutputTemplate, 'string');
+    const firstOutputDir = path.dirname(firstOutputTemplate!);
+    const firstOutputPath = path.join(firstOutputDir, 'media.mkv');
+    fs.mkdirSync(firstOutputDir, { recursive: true });
+    fs.writeFileSync(firstOutputPath, 'cached media');
+    spawnedProcesses[0]?.emit('close', 0);
+
+    await new Promise((resolve) => setImmediate(resolve));
+    assert.equal(await cache.getCachedMediaPath('https://youtu.be/demo'), firstOutputPath);
+
+    cache.start('https://youtu.be/demo', { mode: 'background', maxHeight: 480 });
+
+    assert.equal(spawnCalls.length, 2);
+    assert.equal(fs.existsSync(firstOutputPath), false);
+    assert.equal(await cache.getCachedMediaPath('https://youtu.be/demo'), null);
+    assert.equal(
+      spawnCalls[1]?.args[spawnCalls[1].args.indexOf('-f') + 1],
+      'bestvideo*[height<=480]+bestaudio/best[height<=480]',
+    );
+  } finally {
+    fs.rmSync(cacheRoot, { recursive: true, force: true });
+  }
+});
+
+test('YouTube media cache restarts running sessions when the height cap changes', () => {
+  const cacheRoot = makeTempCacheRoot();
+  const spawnedProcesses: FakeYtDlpProcess[] = [];
+  const spawnCalls: SpawnCall[] = [];
+
+  try {
+    const cache = createYoutubeMediaCacheService({
+      cacheRoot,
+      getYtDlpCommand: () => 'yt-dlp',
+      spawn: (command, args, options) => {
+        spawnCalls.push({ command, args, options });
+        const proc = new FakeYtDlpProcess();
+        spawnedProcesses.push(proc);
+        return proc;
+      },
+    });
+
+    cache.start('https://youtu.be/demo', { mode: 'background', maxHeight: 720 });
+    cache.start('https://youtu.be/demo', { mode: 'background', maxHeight: 480 });
+
+    assert.equal(spawnedProcesses[0]?.killed, true);
+    assert.equal(spawnCalls.length, 2);
+    assert.equal(
+      spawnCalls[1]?.args[spawnCalls[1].args.indexOf('-f') + 1],
+      'bestvideo*[height<=480]+bestaudio/best[height<=480]',
+    );
+  } finally {
+    fs.rmSync(cacheRoot, { recursive: true, force: true });
+  }
+});
+
+test('YouTube media cache removes stale files from previous runs on startup', () => {
+  const cacheRoot = makeTempCacheRoot();
+  const staleDir = path.join(cacheRoot, 'stale-session');
+  const spawnCalls: SpawnCall[] = [];
+
+  try {
+    fs.mkdirSync(staleDir, { recursive: true });
+    fs.writeFileSync(path.join(staleDir, 'media.mkv'), 'stale cached media');
+
+    createYoutubeMediaCacheService({
+      cacheRoot,
+      getYtDlpCommand: () => 'yt-dlp',
+      spawn: (command, args) => {
+        spawnCalls.push({ command, args });
+        return new FakeYtDlpProcess();
+      },
+    });
+
+    assert.equal(fs.existsSync(staleDir), false);
+    assert.deepEqual(spawnCalls, []);
+  } finally {
+    fs.rmSync(cacheRoot, { recursive: true, force: true });
+  }
+});
+
+test('YouTube media cache restarts when a ready cached file was deleted externally', async () => {
+  const cacheRoot = makeTempCacheRoot();
+  const spawnedProcesses: FakeYtDlpProcess[] = [];
+  const spawnCalls: Array<{ command: string; args: string[] }> = [];
+
+  try {
+    const cache = createYoutubeMediaCacheService({
+      cacheRoot,
+      getYtDlpCommand: () => 'yt-dlp',
+      spawn: (command, args) => {
+        spawnCalls.push({ command, args });
+        const proc = new FakeYtDlpProcess();
+        spawnedProcesses.push(proc);
+        return proc;
+      },
+    });
+
+    cache.start('https://youtu.be/demo', { mode: 'background' });
+    const outputTemplate = spawnCalls[0]?.args[spawnCalls[0].args.indexOf('-o') + 1];
+    assert.equal(typeof outputTemplate, 'string');
+    const outputDir = path.dirname(outputTemplate!);
+    const outputPath = path.join(outputDir, 'media.mkv');
+    fs.mkdirSync(outputDir, { recursive: true });
+    fs.writeFileSync(outputPath, 'cached media');
+    spawnedProcesses[0]?.emit('close', 0);
+
+    await new Promise((resolve) => setImmediate(resolve));
+    assert.equal(await cache.getCachedMediaPath('https://youtu.be/demo'), outputPath);
+
+    fs.rmSync(outputPath);
+    cache.start('https://youtu.be/demo', { mode: 'background' });
+
+    assert.equal(spawnCalls.length, 2);
+    assert.equal(await cache.getCachedMediaPath('https://youtu.be/demo'), null);
+  } finally {
+    fs.rmSync(cacheRoot, { recursive: true, force: true });
+  }
+});
+
+test('YouTube media cache removes stale disk siblings before starting a new cache', () => {
+  const cacheRoot = makeTempCacheRoot();
+  const staleDir = path.join(cacheRoot, 'stale-sibling');
+  const spawnCalls: Array<{ command: string; args: string[] }> = [];
+
+  try {
+    const cache = createYoutubeMediaCacheService({
+      cacheRoot,
+      getYtDlpCommand: () => 'yt-dlp',
+      spawn: (command, args) => {
+        spawnCalls.push({ command, args });
+        return new FakeYtDlpProcess();
+      },
+    });
+    fs.mkdirSync(staleDir, { recursive: true });
+    fs.writeFileSync(path.join(staleDir, 'media.mkv'), 'stale cached media');
+
+    cache.start('https://youtu.be/demo', { mode: 'background' });
+
+    assert.equal(fs.existsSync(staleDir), false);
+    assert.equal(spawnCalls.length, 1);
+  } finally {
+    fs.rmSync(cacheRoot, { recursive: true, force: true });
+  }
+});
+
+test('YouTube media cache drops old sessions when a new background cache starts', async () => {
+  const cacheRoot = makeTempCacheRoot();
+  const spawnedProcesses: FakeYtDlpProcess[] = [];
+  const spawnCalls: Array<{ command: string; args: string[] }> = [];
+
+  try {
+    const cache = createYoutubeMediaCacheService({
+      cacheRoot,
+      getYtDlpCommand: () => 'yt-dlp',
+      spawn: (command, args) => {
+        spawnCalls.push({ command, args });
+        const proc = new FakeYtDlpProcess();
+        spawnedProcesses.push(proc);
+        return proc;
+      },
+    });
+
+    cache.start('https://youtu.be/first', { mode: 'background' });
+    const firstOutputTemplate = spawnCalls[0]?.args[spawnCalls[0].args.indexOf('-o') + 1];
+    assert.equal(typeof firstOutputTemplate, 'string');
+    const firstOutputDir = path.dirname(firstOutputTemplate!);
+    fs.mkdirSync(firstOutputDir, { recursive: true });
+    fs.writeFileSync(path.join(firstOutputDir, 'media.mkv'), 'cached media');
+
+    cache.start('https://youtu.be/second', { mode: 'background' });
+
+    assert.equal(spawnedProcesses[0]?.killed, true);
+    assert.equal(fs.existsSync(firstOutputDir), false);
+    assert.equal(await cache.getCachedMediaPath('https://youtu.be/first'), null);
+    assert.equal(spawnCalls.length, 2);
+  } finally {
+    fs.rmSync(cacheRoot, { recursive: true, force: true });
+  }
+});
+
+test('YouTube media cache cleanup kills downloads and removes temp files', async () => {
+  const cacheRoot = makeTempCacheRoot();
+  const spawnedProcesses: FakeYtDlpProcess[] = [];
+  const spawnCalls: Array<{ command: string; args: string[] }> = [];
+
+  try {
+    const cache = createYoutubeMediaCacheService({
+      cacheRoot,
+      getYtDlpCommand: () => 'yt-dlp',
+      spawn: (command, args) => {
+        spawnCalls.push({ command, args });
+        const proc = new FakeYtDlpProcess();
+        spawnedProcesses.push(proc);
+        return proc;
+      },
+    });
+
+    cache.start('https://youtu.be/demo', { mode: 'background' });
+    const outputTemplate = spawnCalls[0]?.args[spawnCalls[0].args.indexOf('-o') + 1];
+    assert.equal(typeof outputTemplate, 'string');
+    const outputDir = path.dirname(outputTemplate!);
+    fs.mkdirSync(outputDir, { recursive: true });
+    fs.writeFileSync(path.join(outputDir, 'media.mkv'), 'cached media');
+
+    cache.cleanup();
+
+    assert.equal(spawnedProcesses[0]?.killed, true);
+    assert.equal(fs.existsSync(outputDir), false);
+    assert.equal(await cache.getCachedMediaPath('https://youtu.be/demo'), null);
+  } finally {
+    fs.rmSync(cacheRoot, { recursive: true, force: true });
+  }
+});

src/core/services/youtube/media-cache.ts

@@ -0,0 +1,307 @@
+import { spawn as spawnProcess } from 'node:child_process';
+import * as crypto from 'node:crypto';
+import { EventEmitter } from 'node:events';
+import * as fs from 'node:fs';
+import * as os from 'node:os';
+import * as path from 'node:path';
+
+import type { YoutubeMediaCacheMode } from '../../../types/integrations';
+import { getYoutubeYtDlpCommand } from './ytdlp-command';
+
+type MediaCacheSessionState = 'running' | 'ready' | 'failed';
+
+type SpawnedProcess = EventEmitter & {
+  killed?: boolean;
+  kill?: (signal?: NodeJS.Signals | number) => boolean;
+};
+
+type SpawnProcess = (
+  command: string,
+  args: string[],
+  options?: { stdio?: Array<'ignore' | 'pipe'> },
+) => SpawnedProcess;
+
+interface MediaCacheSession {
+  url: string;
+  dir: string;
+  maxHeight: number;
+  process: SpawnedProcess | null;
+  readyPath: string | null;
+  state: MediaCacheSessionState;
+}
+
+export interface YoutubeMediaCacheStartOptions {
+  mode: YoutubeMediaCacheMode;
+  maxHeight?: number;
+}
+
+export interface YoutubeMediaCacheServiceDeps {
+  cacheRoot?: string;
+  getYtDlpCommand?: () => string;
+  spawn?: SpawnProcess;
+  onDownloadStarted?: (event: { url: string }) => void;
+  onReady?: (event: { url: string; path: string }) => void;
+  onFailed?: (event: { url: string }) => void;
+  logInfo?: (message: string) => void;
+  logWarn?: (message: string) => void;
+}
+
+const MEDIA_FILE_EXTENSIONS = new Set(['.mkv', '.mp4', '.webm', '.m4a', '.mp3', '.opus']);
+const DEFAULT_MAX_HEIGHT = 720;
+
+function cacheKeyForUrl(url: string): string {
+  return crypto.createHash('sha256').update(url).digest('hex').slice(0, 24);
+}
+
+function isFinalMediaFile(fileName: string): boolean {
+  if (!fileName.startsWith('media.')) {
+    return false;
+  }
+  if (fileName.endsWith('.part') || fileName.endsWith('.ytdl') || fileName.endsWith('.tmp')) {
+    return false;
+  }
+  return MEDIA_FILE_EXTENSIONS.has(path.extname(fileName).toLowerCase());
+}
+
+function findReadyMediaPath(dir: string): string | null {
+  try {
+    const files = fs.readdirSync(dir);
+    const mediaFile = files.find(isFinalMediaFile);
+    return mediaFile ? path.join(dir, mediaFile) : null;
+  } catch {
+    return null;
+  }
+}
+
+function getFormatSelector(maxHeight: number): string {
+  return maxHeight > 0
+    ? `bestvideo*[height<=${maxHeight}]+bestaudio/best[height<=${maxHeight}]`
+    : 'bestvideo*+bestaudio/best';
+}
+
+function normalizeMaxHeight(maxHeight: number | undefined): number {
+  if (maxHeight === undefined) {
+    return DEFAULT_MAX_HEIGHT;
+  }
+  return Number.isInteger(maxHeight) && maxHeight >= 0 ? maxHeight : DEFAULT_MAX_HEIGHT;
+}
+
+function createYtDlpArgs(url: string, outputTemplate: string, maxHeight?: number): string[] {
+  return [
+    '--no-playlist',
+    '--no-warnings',
+    '--force-ipv4',
+    '--retries',
+    '5',
+    '--fragment-retries',
+    '5',
+    '--extractor-retries',
+    '5',
+    '-f',
+    getFormatSelector(normalizeMaxHeight(maxHeight)),
+    '--merge-output-format',
+    'mkv',
+    '-o',
+    outputTemplate,
+    url,
+  ];
+}
+
+export function createYoutubeMediaCacheService(deps: YoutubeMediaCacheServiceDeps = {}) {
+  const cacheRoot = deps.cacheRoot ?? path.join(os.tmpdir(), 'subminer-youtube-media-cache');
+  const getYtDlpCommand = deps.getYtDlpCommand ?? getYoutubeYtDlpCommand;
+  const spawn: SpawnProcess =
+    deps.spawn ??
+    ((command, args, options) =>
+      spawnProcess(command, args, options ?? {}) as unknown as SpawnedProcess);
+  const sessions = new Map<string, MediaCacheSession>();
+  let activeKey: string | null = null;
+
+  const getSessionDir = (url: string): string => path.join(cacheRoot, cacheKeyForUrl(url));
+  const removeCacheDir = (dir: string): void => {
+    try {
+      fs.rmSync(dir, { recursive: true, force: true });
+    } catch {
+      // Temp cache cleanup should not block shutdown or playback startup.
+    }
+  };
+  const removeCacheRootEntriesExcept = (dirsToKeep: string[]): void => {
+    let entries: fs.Dirent[];
+    try {
+      entries = fs.readdirSync(cacheRoot, { withFileTypes: true });
+    } catch {
+      return;
+    }
+    const keepDirs = new Set(dirsToKeep.map((dir) => path.resolve(dir)));
+    for (const entry of entries) {
+      const entryPath = path.join(cacheRoot, entry.name);
+      if (keepDirs.has(path.resolve(entryPath))) {
+        continue;
+      }
+      removeCacheDir(entryPath);
+    }
+  };
+  const removeSession = (key: string): void => {
+    const session = sessions.get(key);
+    if (!session) {
+      return;
+    }
+    if (session.state === 'running' && session.process?.kill && !session.process.killed) {
+      session.process.kill();
+    }
+    sessions.delete(key);
+    if (activeKey === key) {
+      activeKey = null;
+    }
+    removeCacheDir(session.dir);
+  };
+  const removeInactiveSessions = (keyToKeep: string): void => {
+    for (const key of [...sessions.keys()]) {
+      if (key !== keyToKeep) {
+        removeSession(key);
+      }
+    }
+  };
+  removeCacheRootEntriesExcept([]);
+
+  const getCachedMediaPath = async (url: string): Promise<string | null> => {
+    const key = cacheKeyForUrl(url);
+    const session = sessions.get(key);
+    if (session?.readyPath && fs.existsSync(session.readyPath)) {
+      return session.readyPath;
+    }
+
+    const readyPath = findReadyMediaPath(session?.dir ?? getSessionDir(url));
+    if (readyPath) {
+      sessions.set(key, {
+        url,
+        dir: path.dirname(readyPath),
+        maxHeight: session?.maxHeight ?? DEFAULT_MAX_HEIGHT,
+        process: null,
+        readyPath,
+        state: 'ready',
+      });
+      return readyPath;
+    }
+
+    return null;
+  };
+
+  const getActiveCachedMediaPath = async (): Promise<string | null> => {
+    if (!activeKey) {
+      return null;
+    }
+    const session = sessions.get(activeKey);
+    return session ? getCachedMediaPath(session.url) : null;
+  };
+
+  const start = (url: string, options: YoutubeMediaCacheStartOptions): void => {
+    if (options.mode !== 'background') {
+      if (activeKey) {
+        const activeSession = sessions.get(activeKey);
+        if (activeSession?.state === 'running') {
+          removeSession(activeKey);
+        }
+      }
+      activeKey = null;
+      return;
+    }
+
+    const key = cacheKeyForUrl(url);
+    const maxHeight = normalizeMaxHeight(options.maxHeight);
+    activeKey = key;
+    const dir = getSessionDir(url);
+    const existingSession = sessions.get(key);
+    const canReuseExistingSession = existingSession?.maxHeight === maxHeight;
+    if (existingSession?.state === 'running' && canReuseExistingSession) {
+      removeInactiveSessions(key);
+      removeCacheRootEntriesExcept([existingSession.dir]);
+      return;
+    }
+    if (existingSession) {
+      if (
+        canReuseExistingSession &&
+        existingSession.state === 'ready' &&
+        ((existingSession.readyPath && fs.existsSync(existingSession.readyPath)) ||
+          findReadyMediaPath(existingSession.dir))
+      ) {
+        removeInactiveSessions(key);
+        removeCacheRootEntriesExcept([existingSession.dir]);
+        return;
+      }
+      removeSession(key);
+      activeKey = key;
+    }
+    removeInactiveSessions(key);
+    removeCacheRootEntriesExcept([dir]);
+
+    fs.mkdirSync(dir, { recursive: true });
+    const outputTemplate = path.join(dir, 'media.%(ext)s');
+    const args = createYtDlpArgs(url, outputTemplate, maxHeight);
+    const child = spawn(getYtDlpCommand(), args, { stdio: ['ignore', 'ignore', 'ignore'] });
+    const session: MediaCacheSession = {
+      url,
+      dir,
+      maxHeight,
+      process: child,
+      readyPath: null,
+      state: 'running',
+    };
+    sessions.set(key, session);
+    deps.logInfo?.(`Started YouTube media cache download for ${url}`);
+    deps.onDownloadStarted?.({ url });
+
+    child.once('error', (error) => {
+      const currentSession = sessions.get(key);
+      if (currentSession !== session) {
+        return;
+      }
+      session.state = 'failed';
+      session.process = null;
+      deps.logWarn?.(
+        `YouTube media cache download failed: ${
+          error instanceof Error ? error.message : String(error)
+        }`,
+      );
+      deps.onFailed?.({ url });
+    });
+
+    child.once('close', (code) => {
+      const currentSession = sessions.get(key);
+      if (currentSession !== session) {
+        return;
+      }
+      if (session.state === 'failed') {
+        return;
+      }
+      session.process = null;
+      if (code === 0) {
+        const readyPath = findReadyMediaPath(dir);
+        if (readyPath) {
+          session.state = 'ready';
+          session.readyPath = readyPath;
+          deps.logInfo?.(`YouTube media cache ready at ${readyPath}`);
+          deps.onReady?.({ url, path: readyPath });
+          return;
+        }
+      }
+      session.state = 'failed';
+      deps.logWarn?.(`YouTube media cache download exited without a usable media file.`);
+      deps.onFailed?.({ url });
+    });
+  };
+
+  const cleanup = (): void => {
+    for (const key of [...sessions.keys()]) {
+      removeSession(key);
+    }
+    activeKey = null;
+  };
+
+  return {
+    cleanup,
+    getActiveCachedMediaPath,
+    getCachedMediaPath,
+    start,
+  };
+}

src/main.ts

@@ -332,6 +332,7 @@ import {
   acquireYoutubeSubtitleTrack,
   acquireYoutubeSubtitleTracks,
 } from './core/services/youtube/generate';
+import { createYoutubeMediaCacheService } from './core/services/youtube/media-cache';
 import { resolveYoutubePlaybackUrl } from './core/services/youtube/playback-resolve';
 import { probeYoutubeTracks } from './core/services/youtube/track-probe';
 import {
@@ -1172,6 +1173,53 @@ const prepareYoutubePlaybackInMpv = createPrepareYoutubePlaybackInMpvHandler({
   },
   wait: (ms) => new Promise((resolve) => setTimeout(resolve, ms)),
 });
+const youtubeMediaCache = createYoutubeMediaCacheService({
+  onDownloadStarted: (event) => {
+    showConfiguredStatusNotification('YouTube media cache is downloading.', {
+      id: 'youtube-media-cache-status',
+      title: 'YouTube media cache',
+      variant: 'progress',
+      persistent: true,
+    });
+    logger.info(`YouTube media cache download notification shown for ${event.url}`);
+  },
+  onReady: (event) => {
+    showConfiguredStatusNotification('YouTube media cache ready.', {
+      id: 'youtube-media-cache-status',
+      title: 'YouTube media cache',
+      variant: 'success',
+      persistent: false,
+    });
+    void appState.ankiIntegration
+      ?.handleYoutubeMediaCacheReady(event.url, event.path, { notifyNoQueued: false })
+      .catch((error) => {
+        logger.warn(
+          `Failed to apply queued YouTube media updates: ${
+            error instanceof Error ? error.message : String(error)
+          }`,
+        );
+      });
+  },
+  onFailed: (event) => {
+    showConfiguredStatusNotification('YouTube media cache failed.', {
+      id: 'youtube-media-cache-status',
+      title: 'YouTube media cache',
+      variant: 'error',
+      persistent: false,
+    });
+    void appState.ankiIntegration
+      ?.handleYoutubeMediaCacheFailed(event.url, { notifyStatus: false })
+      .catch((error) => {
+        logger.warn(
+          `Failed to drain queued YouTube media updates after cache failure: ${
+            error instanceof Error ? error.message : String(error)
+          }`,
+        );
+      });
+  },
+  logInfo: (message) => logger.info(message),
+  logWarn: (message) => logger.warn(message),
+});
 const waitForYoutubeMpvConnected = createWaitForMpvConnectedHandler({
   getMpvClient: () => appState.mpvClient,
   now: () => Date.now(),
@@ -1316,6 +1364,12 @@ const youtubePlaybackRuntime = createYoutubePlaybackRuntime({
   },
   waitForYoutubeMpvConnected: (timeoutMs) => waitForYoutubeMpvConnected(timeoutMs),
   prepareYoutubePlaybackInMpv: (request) => prepareYoutubePlaybackInMpv(request),
+  startYoutubeMediaCache: (url) => {
+    youtubeMediaCache.start(url, {
+      mode: getResolvedConfig().youtube.mediaCache.mode,
+      maxHeight: getResolvedConfig().youtube.mediaCache.maxHeight,
+    });
+  },
   runYoutubePlaybackFlow: (request) => youtubeFlowRuntime.runYoutubePlaybackFlow(request),
   logInfo: (message) => logger.info(message),
   logWarn: (message) => logger.warn(message),
@@ -1635,6 +1689,30 @@ function isYoutubePlaybackActiveNow(): boolean {
   );
 }
 
+function shouldRequireYoutubeMediaCacheForCurrentPlayback(): boolean {
+  return (
+    getResolvedConfig().youtube.mediaCache.mode === 'background' && isYoutubePlaybackActiveNow()
+  );
+}
+
+async function getCachedYoutubeMediaPathForCurrentPlayback(
+  currentVideoPath: string,
+  _kind: 'audio' | 'video',
+): Promise<string | null> {
+  if (getResolvedConfig().youtube.mediaCache.mode !== 'background') {
+    return null;
+  }
+  if (!isYoutubePlaybackActiveNow()) {
+    return null;
+  }
+  // mpv can expose the resolved stream URL here while the cache key uses the original page URL.
+  // Keep the active-cache fallback so current playback can still resolve the ready cached file.
+  return (
+    (await youtubeMediaCache.getCachedMediaPath(currentVideoPath)) ??
+    (await youtubeMediaCache.getActiveCachedMediaPath())
+  );
+}
+
 function reportYoutubeSubtitleFailure(message: string): void {
   const type = getConfiguredStatusNotificationType();
   if (type === 'none') {
@@ -2636,6 +2714,7 @@ const overlayNotificationsRuntime = createOverlayNotificationsRuntime({
 });
 const {
   flushQueuedOverlayNotifications,
+  flushQueuedMpvOsdNotifications,
   openAnkiCardFromNotification,
   toggleNotificationHistoryPanel,
   showConfiguredPlaybackFeedback,
@@ -3801,6 +3880,7 @@ const {
     },
     stopJellyfinRemoteSession: () => stopJellyfinRemoteSession(),
     cleanupYoutubeSubtitleTempDirs: () => youtubeFlowRuntime.cleanupSubtitleTempDirs(),
+    cleanupYoutubeMediaCache: () => youtubeMediaCache.cleanup(),
     cleanupJellyfinSubtitleCache: () => cleanupJellyfinSubtitleCache(),
     stopDiscordPresenceService: () => {
       void appState.discordPresenceService?.stop();
@@ -4286,6 +4366,7 @@ const {
     },
     onMpvConnected: () => {
       maybeStartOverlayLoadingOsd();
+      flushQueuedMpvOsdNotifications();
       if (appState.sessionBindingsInitialized) {
         sendMpvCommandRuntime(appState.mpvClient, [
           'script-message',
@@ -5731,6 +5812,9 @@ const { registerIpcRuntimeHandlers } = composeIpcRuntimeHandlers({
         );
       },
       getKnownWordCacheStatePath: () => path.join(USER_DATA_PATH, 'known-words-cache.json'),
+      getCachedMediaPath: (currentVideoPath, kind) =>
+        getCachedYoutubeMediaPathForCurrentPlayback(currentVideoPath, kind),
+      shouldRequireRemoteMediaCache: () => shouldRequireYoutubeMediaCacheForCurrentPlayback(),
       showDesktopNotification,
       showOverlayNotification,
       createFieldGroupingCallback: () => createFieldGroupingCallback(),
@@ -6207,6 +6291,9 @@ const { initializeOverlayRuntime: initializeOverlayRuntimeHandler } =
       showOverlayNotification,
       createFieldGroupingCallback: () => createFieldGroupingCallback(),
       getKnownWordCacheStatePath: () => path.join(USER_DATA_PATH, 'known-words-cache.json'),
+      getCachedMediaPath: (currentVideoPath, kind) =>
+        getCachedYoutubeMediaPathForCurrentPlayback(currentVideoPath, kind),
+      shouldRequireRemoteMediaCache: () => shouldRequireYoutubeMediaCacheForCurrentPlayback(),
       shouldStartAnkiIntegration: () =>
         !(appState.initialArgs && isHeadlessInitialCommand(appState.initialArgs)),
     },

src/main/dependencies.ts

@@ -126,6 +126,8 @@ export interface AnkiJimakuIpcRuntimeServiceDepsParams {
   getAnkiIntegration: AnkiJimakuIpcRuntimeOptions['getAnkiIntegration'];
   setAnkiIntegration: AnkiJimakuIpcRuntimeOptions['setAnkiIntegration'];
   getKnownWordCacheStatePath: AnkiJimakuIpcRuntimeOptions['getKnownWordCacheStatePath'];
+  getCachedMediaPath?: AnkiJimakuIpcRuntimeOptions['getCachedMediaPath'];
+  shouldRequireRemoteMediaCache?: AnkiJimakuIpcRuntimeOptions['shouldRequireRemoteMediaCache'];
   showDesktopNotification: AnkiJimakuIpcRuntimeOptions['showDesktopNotification'];
   showOverlayNotification?: (payload: OverlayNotificationPayload) => void;
   createFieldGroupingCallback: AnkiJimakuIpcRuntimeOptions['createFieldGroupingCallback'];
@@ -317,6 +319,10 @@ export function createAnkiJimakuIpcRuntimeServiceDeps(
     getAnkiIntegration: params.getAnkiIntegration,
     setAnkiIntegration: params.setAnkiIntegration,
     getKnownWordCacheStatePath: params.getKnownWordCacheStatePath,
+    ...(params.getCachedMediaPath ? { getCachedMediaPath: params.getCachedMediaPath } : {}),
+    ...(params.shouldRequireRemoteMediaCache
+      ? { shouldRequireRemoteMediaCache: params.shouldRequireRemoteMediaCache }
+      : {}),
     showDesktopNotification: params.showDesktopNotification,
     showOverlayNotification: params.showOverlayNotification,
     createFieldGroupingCallback: params.createFieldGroupingCallback,

src/main/main-wiring.test.ts

@@ -514,6 +514,48 @@ test('configured overlay notifications require visible ready overlay window', ()
   assert.match(statusBlock, /isOverlayReady: \(\) => isVisibleOverlayContentReady\(\)/);
 });
 
+test('YouTube media cache lifecycle routes through configured status notifications', () => {
+  const source = readMainSource();
+  const cacheBlock = source.match(
+    /const youtubeMediaCache = createYoutubeMediaCacheService\(\{(?<body>[\s\S]*?)\n\}\);\nconst waitForYoutubeMpvConnected/,
+  )?.groups?.body;
+  const startCacheBlock = source.match(
+    /startYoutubeMediaCache:\s*\(url\)\s*=>\s*\{(?<body>[\s\S]*?)\n  \},\n  runYoutubePlaybackFlow/,
+  )?.groups?.body;
+
+  assert.ok(cacheBlock);
+  assert.ok(startCacheBlock);
+  assert.match(
+    cacheBlock,
+    /onDownloadStarted:\s*\(event\)\s*=>\s*\{[\s\S]*showConfiguredStatusNotification\(\s*'YouTube media cache is downloading\.'/,
+  );
+  assert.match(cacheBlock, /id:\s*'youtube-media-cache-status'/);
+  assert.match(cacheBlock, /variant:\s*'progress'/);
+  assert.match(cacheBlock, /persistent:\s*true/);
+  assert.match(
+    cacheBlock,
+    /onReady:\s*\(event\)\s*=>\s*\{[\s\S]*showConfiguredStatusNotification\(\s*'YouTube media cache ready\.'/,
+  );
+  assert.match(cacheBlock, /variant:\s*'success'/);
+  assert.match(cacheBlock, /notifyNoQueued:\s*false/);
+  assert.match(startCacheBlock, /mode:\s*getResolvedConfig\(\)\.youtube\.mediaCache\.mode/);
+  assert.match(
+    startCacheBlock,
+    /maxHeight:\s*getResolvedConfig\(\)\.youtube\.mediaCache\.maxHeight/,
+  );
+});
+
+test('mpv connection flushes queued configured OSD notifications', () => {
+  const source = readMainSource();
+  const connectedBlock = source.match(
+    /onMpvConnected:\s*\(\)\s*=>\s*\{(?<body>[\s\S]*?)\n    \},\n    maybeRunAnilistPostWatchUpdate:/,
+  )?.groups?.body;
+
+  assert.ok(connectedBlock);
+  assert.match(source, /flushQueuedMpvOsdNotifications/);
+  assert.match(connectedBlock, /flushQueuedMpvOsdNotifications\(\);/);
+});
+
 test('manual visible overlay show primes current subtitle from mpv before relying on live events', () => {
   const source = readMainSource();
   const setBlock = source.match(

src/main/runtime/app-lifecycle-actions.test.ts

@@ -41,18 +41,20 @@ test('on will quit cleanup handler runs all cleanup steps', () => {
     clearYomitanSettingsWindow: () => calls.push('clear-yomitan-settings-window'),
     stopJellyfinRemoteSession: () => calls.push('stop-jellyfin-remote'),
     cleanupYoutubeSubtitleTempDirs: () => calls.push('cleanup-youtube-subtitles'),
+    cleanupYoutubeMediaCache: () => calls.push('cleanup-youtube-media'),
     cleanupJellyfinSubtitleCache: () => calls.push('cleanup-jellyfin-subtitles'),
     stopDiscordPresenceService: () => calls.push('stop-discord-presence'),
   });
 
   cleanup();
-  assert.equal(calls.length, 32);
+  assert.equal(calls.length, 33);
   assert.equal(calls[0], 'destroy-tray');
   assert.equal(calls[calls.length - 1], 'stop-discord-presence');
   assert.ok(calls.includes('cleanup-jellyfin-subtitles'));
   assert.ok(calls.includes('clear-windows-visible-overlay-poll'));
   assert.ok(calls.includes('clear-linux-mpv-fullscreen-overlay-refresh-timeouts'));
   assert.ok(calls.includes('cleanup-youtube-subtitles'));
+  assert.ok(calls.includes('cleanup-youtube-media'));
   assert.ok(calls.indexOf('flush-mpv-log') < calls.indexOf('destroy-socket'));
 });
 
@@ -92,6 +94,7 @@ test('on will quit cleanup handler cleans jellyfin subtitle cache when stopping
       throw new Error('stop failed');
     },
     cleanupYoutubeSubtitleTempDirs: () => calls.push('cleanup-youtube-subtitles'),
+    cleanupYoutubeMediaCache: () => calls.push('cleanup-youtube-media'),
     cleanupJellyfinSubtitleCache: () => calls.push('cleanup-jellyfin-subtitles'),
     stopDiscordPresenceService: () => calls.push('stop-discord-presence'),
   });

src/main/runtime/app-lifecycle-actions.ts

@@ -29,6 +29,7 @@ export function createOnWillQuitCleanupHandler(deps: {
   clearYomitanSettingsWindow: () => void;
   stopJellyfinRemoteSession: () => void;
   cleanupYoutubeSubtitleTempDirs: () => void;
+  cleanupYoutubeMediaCache: () => void;
   cleanupJellyfinSubtitleCache: () => void;
   stopDiscordPresenceService: () => void;
 }) {
@@ -67,6 +68,7 @@ export function createOnWillQuitCleanupHandler(deps: {
       deps.cleanupJellyfinSubtitleCache();
     }
     deps.cleanupYoutubeSubtitleTempDirs();
+    deps.cleanupYoutubeMediaCache();
     deps.stopDiscordPresenceService();
   };
 }

src/main/runtime/app-lifecycle-main-cleanup.test.ts

@@ -70,6 +70,7 @@ test('cleanup deps builder returns handlers that guard optional runtime objects'
 
     stopJellyfinRemoteSession: () => calls.push('stop-jellyfin-remote'),
     cleanupYoutubeSubtitleTempDirs: () => calls.push('cleanup-youtube-subtitles'),
+    cleanupYoutubeMediaCache: () => calls.push('cleanup-youtube-media'),
     cleanupJellyfinSubtitleCache: () => calls.push('cleanup-jellyfin-subtitles'),
     stopDiscordPresenceService: () => calls.push('stop-discord-presence'),
   });
@@ -92,6 +93,7 @@ test('cleanup deps builder returns handlers that guard optional runtime objects'
   assert.ok(calls.includes('destroy-yomitan-settings-window'));
   assert.ok(calls.includes('stop-jellyfin-remote'));
   assert.ok(calls.includes('cleanup-youtube-subtitles'));
+  assert.ok(calls.includes('cleanup-youtube-media'));
   assert.ok(calls.includes('cleanup-jellyfin-subtitles'));
   assert.ok(calls.includes('stop-discord-presence'));
   assert.ok(calls.includes('clear-windows-visible-overlay-foreground-poll-loop'));
@@ -147,6 +149,7 @@ test('cleanup deps builder skips destroyed yomitan window', () => {
     clearYomitanSettingsWindow: () => {},
     stopJellyfinRemoteSession: () => {},
     cleanupYoutubeSubtitleTempDirs: () => {},
+    cleanupYoutubeMediaCache: () => {},
     cleanupJellyfinSubtitleCache: () => {},
     stopDiscordPresenceService: () => {},
   });
@@ -197,6 +200,7 @@ test('cleanup deps builder skips global shortcut cleanup before app ready', () =
     clearYomitanSettingsWindow: () => {},
     stopJellyfinRemoteSession: () => {},
     cleanupYoutubeSubtitleTempDirs: () => {},
+    cleanupYoutubeMediaCache: () => {},
     cleanupJellyfinSubtitleCache: () => {},
     stopDiscordPresenceService: () => {},
   });

src/main/runtime/app-lifecycle-main-cleanup.ts

@@ -58,6 +58,7 @@ export function createBuildOnWillQuitCleanupDepsHandler(deps: {
 
   stopJellyfinRemoteSession: () => void;
   cleanupYoutubeSubtitleTempDirs: () => void;
+  cleanupYoutubeMediaCache: () => void;
   cleanupJellyfinSubtitleCache: () => void;
   stopDiscordPresenceService: () => void;
 }) {
@@ -142,6 +143,7 @@ export function createBuildOnWillQuitCleanupDepsHandler(deps: {
     clearYomitanSettingsWindow: () => deps.clearYomitanSettingsWindow(),
     stopJellyfinRemoteSession: () => deps.stopJellyfinRemoteSession(),
     cleanupYoutubeSubtitleTempDirs: () => deps.cleanupYoutubeSubtitleTempDirs(),
+    cleanupYoutubeMediaCache: () => deps.cleanupYoutubeMediaCache(),
     cleanupJellyfinSubtitleCache: () => deps.cleanupJellyfinSubtitleCache(),
     stopDiscordPresenceService: () => deps.stopDiscordPresenceService(),
   });

src/main/runtime/composers/startup-lifecycle-composer.test.ts

@@ -49,6 +49,7 @@ test('composeStartupLifecycleHandlers returns callable startup lifecycle handler
       clearYomitanSettingsWindow: () => {},
       stopJellyfinRemoteSession: async () => {},
       cleanupYoutubeSubtitleTempDirs: () => {},
+      cleanupYoutubeMediaCache: () => {},
       cleanupJellyfinSubtitleCache: () => {},
       stopDiscordPresenceService: () => {},
     },

src/main/runtime/configured-status-notification.test.ts

@@ -28,7 +28,7 @@ test('notifyConfiguredStatus routes both to overlay and system without osd', ()
   ]);
 });
 
-test('notifyConfiguredStatus falls back to desktop for pre-overlay both status', () => {
+test('notifyConfiguredStatus queues overlay for pre-overlay both status and preserves desktop', () => {
   const calls: string[] = [];
 
   notifyConfiguredStatus('Overlay loading...', {
@@ -43,10 +43,10 @@ test('notifyConfiguredStatus falls back to desktop for pre-overlay both status',
       calls.push(`desktop:${title}:${options.body ?? ''}`),
   });
 
-  assert.deepEqual(calls, ['desktop:SubMiner:Overlay loading...']);
+  assert.deepEqual(calls, ['overlay::Overlay loading...', 'desktop:SubMiner:Overlay loading...']);
 });
 
-test('notifyConfiguredStatus falls back to desktop for pre-overlay overlay-only status', () => {
+test('notifyConfiguredStatus queues overlay for pre-overlay overlay-only status', () => {
   const calls: string[] = [];
 
   notifyConfiguredStatus('Overlay loading...', {
@@ -61,7 +61,7 @@ test('notifyConfiguredStatus falls back to desktop for pre-overlay overlay-only
       calls.push(`desktop:${title}:${options.body ?? ''}`),
   });
 
-  assert.deepEqual(calls, ['desktop:SubMiner:Overlay loading...']);
+  assert.deepEqual(calls, ['overlay::Overlay loading...']);
 });
 
 test('notifyConfiguredStatus routes pre-overlay system status to desktop only', () => {
@@ -97,6 +97,37 @@ test('notifyConfiguredStatus keeps osd-system on legacy surfaces', () => {
   assert.deepEqual(calls, ['osd:Overlay loading...', 'desktop:SubMiner:Overlay loading...']);
 });
 
+test('notifyConfiguredStatus queues osd status when mpv osd is unavailable', () => {
+  const calls: string[] = [];
+
+  notifyConfiguredStatus(
+    'YouTube media cache is downloading.',
+    {
+      getNotificationType: () => 'osd',
+      showOsd: (message) => {
+        calls.push(`osd:${message}`);
+        return false;
+      },
+      queueOsd: (message, options) => {
+        calls.push(`queue:${options.id ?? ''}:${message}`);
+      },
+      showDesktopNotification: (title, options) =>
+        calls.push(`desktop:${title}:${options.body ?? ''}`),
+    },
+    {
+      id: 'youtube-media-cache-status',
+      title: 'YouTube media cache',
+      variant: 'progress',
+      persistent: true,
+    },
+  );
+
+  assert.deepEqual(calls, [
+    'osd:YouTube media cache is downloading.',
+    'queue:youtube-media-cache-status:YouTube media cache is downloading.',
+  ]);
+});
+
 test('notifyConfiguredStatus can suppress desktop delivery for progress ticks', () => {
   const calls: string[] = [];
 

src/main/runtime/configured-status-notification.ts

@@ -5,6 +5,7 @@ export interface ConfiguredStatusNotificationDeps {
   getNotificationType: () => NotificationType | undefined;
   isOverlayReady?: () => boolean;
   showOsd: (message: string) => boolean | void;
+  queueOsd?: (message: string, options: ConfiguredStatusNotificationOptions) => void;
   showOverlayNotification?: (payload: OverlayNotificationPayload) => void;
   showDesktopNotification: (title: string, options: { body?: string }) => void;
 }
@@ -50,8 +51,7 @@ export function notifyConfiguredStatus(
   }
 
   if (showOverlay) {
-    const overlayReady = deps.isOverlayReady?.() ?? true;
-    if (deps.showOverlayNotification && overlayReady) {
+    if (deps.showOverlayNotification) {
       deps.showOverlayNotification({
         id: options.id,
         title: options.title ?? 'SubMiner',
@@ -65,7 +65,10 @@ export function notifyConfiguredStatus(
   }
 
   if (showOsd) {
-    deps.showOsd(message);
+    const shown = deps.showOsd(message);
+    if (shown === false && delivery !== 'feedback') {
+      deps.queueOsd?.(message, options);
+    }
   }
 
   if (desktopEnabled && shouldShowDesktop(type)) {

src/main/runtime/overlay-notifications-runtime.ts

@@ -32,7 +32,7 @@ export interface OverlayNotificationsRuntimeDeps {
   getMainOverlayWindow: () => BrowserWindow | null;
   getVisibleOverlayVisible: () => boolean;
   broadcastToOverlayWindows: (channel: string, ...args: unknown[]) => void;
-  showMpvOsd: (message: string) => void;
+  showMpvOsd: (message: string) => boolean | void;
   getMpvClient: () => MpvIpcClient | null;
   getAnkiIntegration: () => AnkiIntegration | null;
   getRuntimeOptionsManager: () => RuntimeOptionsManager | null;
@@ -42,6 +42,7 @@ export function createOverlayNotificationsRuntime(deps: OverlayNotificationsRunt
   isVisibleOverlayContentReady: () => boolean;
   getConfiguredStatusNotificationType: () => NotificationType;
   flushQueuedOverlayNotifications: () => void;
+  flushQueuedMpvOsdNotifications: () => void;
   showOverlayNotification: (payload: OverlayNotificationPayload) => void;
   dismissOverlayNotification: (id: string) => void;
   openAnkiCardFromNotification: (noteId: number) => Promise<void>;
@@ -95,11 +96,38 @@ export function createOverlayNotificationsRuntime(deps: OverlayNotificationsRunt
   });
   let overlayLoadingOsdController: ReturnType<typeof createOverlayLoadingOsdController> | null =
     null;
+  const queuedConfiguredOsdNotifications = new Map<
+    string,
+    { message: string; options: ConfiguredStatusNotificationOptions }
+  >();
 
   function flushQueuedOverlayNotifications(): void {
     overlayNotificationDelivery.flush();
   }
 
+  function queueConfiguredOsdNotification(
+    message: string,
+    options: ConfiguredStatusNotificationOptions,
+  ): void {
+    const key = options.id ?? message;
+    queuedConfiguredOsdNotifications.set(key, { message, options });
+    while (queuedConfiguredOsdNotifications.size > 16) {
+      const oldestKey = queuedConfiguredOsdNotifications.keys().next().value;
+      if (typeof oldestKey !== 'string') {
+        break;
+      }
+      queuedConfiguredOsdNotifications.delete(oldestKey);
+    }
+  }
+
+  function flushQueuedMpvOsdNotifications(): void {
+    for (const [key, entry] of [...queuedConfiguredOsdNotifications.entries()]) {
+      if (deps.showMpvOsd(entry.message) !== false) {
+        queuedConfiguredOsdNotifications.delete(key);
+      }
+    }
+  }
+
   function sendOverlayNotificationEvent(payload: OverlayNotificationEventPayload): void {
     overlayNotificationDelivery.send(payload);
   }
@@ -145,6 +173,7 @@ export function createOverlayNotificationsRuntime(deps: OverlayNotificationsRunt
         getNotificationType: () => deps.getResolvedConfig().ankiConnect.behavior.notificationType,
         isOverlayReady: () => isVisibleOverlayContentReady(),
         showOsd: (text) => deps.showMpvOsd(text),
+        queueOsd: (text, queueOptions) => queueConfiguredOsdNotification(text, queueOptions),
         showOverlayNotification,
         showDesktopNotification: (title, notificationOptions) =>
           showDesktopNotification(title, notificationOptions),
@@ -238,6 +267,7 @@ export function createOverlayNotificationsRuntime(deps: OverlayNotificationsRunt
     isVisibleOverlayContentReady,
     getConfiguredStatusNotificationType,
     flushQueuedOverlayNotifications,
+    flushQueuedMpvOsdNotifications,
     showOverlayNotification,
     dismissOverlayNotification,
     openAnkiCardFromNotification,

src/main/runtime/overlay-runtime-options-main-deps.ts

@@ -41,6 +41,8 @@ export function createBuildInitializeOverlayRuntimeMainDepsHandler(deps: {
   showOverlayNotification?: (payload: OverlayNotificationPayload) => void;
   createFieldGroupingCallback: OverlayRuntimeOptionsMainDeps['createFieldGroupingCallback'];
   getKnownWordCacheStatePath: () => string;
+  getCachedMediaPath?: OverlayRuntimeOptionsMainDeps['getCachedMediaPath'];
+  shouldRequireRemoteMediaCache?: OverlayRuntimeOptionsMainDeps['shouldRequireRemoteMediaCache'];
   shouldStartAnkiIntegration: () => boolean;
   bindOverlayOwner?: () => void;
   releaseOverlayOwner?: () => void;
@@ -77,6 +79,10 @@ export function createBuildInitializeOverlayRuntimeMainDepsHandler(deps: {
     showOverlayNotification: deps.showOverlayNotification,
     createFieldGroupingCallback: () => deps.createFieldGroupingCallback(),
     getKnownWordCacheStatePath: () => deps.getKnownWordCacheStatePath(),
+    ...(deps.getCachedMediaPath ? { getCachedMediaPath: deps.getCachedMediaPath } : {}),
+    ...(deps.shouldRequireRemoteMediaCache
+      ? { shouldRequireRemoteMediaCache: deps.shouldRequireRemoteMediaCache }
+      : {}),
     shouldStartAnkiIntegration: () => deps.shouldStartAnkiIntegration(),
     bindOverlayOwner: deps.bindOverlayOwner,
     releaseOverlayOwner: deps.releaseOverlayOwner,

src/main/runtime/overlay-runtime-options.ts

@@ -37,6 +37,11 @@ type OverlayRuntimeOptions = {
     data: KikuFieldGroupingRequestData,
   ) => Promise<KikuFieldGroupingChoice>;
   getKnownWordCacheStatePath: () => string;
+  getCachedMediaPath?: (
+    currentVideoPath: string,
+    kind: 'audio' | 'video',
+  ) => Promise<string | null>;
+  shouldRequireRemoteMediaCache?: () => boolean;
   shouldStartAnkiIntegration: () => boolean;
   bindOverlayOwner?: () => void;
   releaseOverlayOwner?: () => void;
@@ -71,6 +76,11 @@ export function createBuildInitializeOverlayRuntimeOptionsHandler(deps: {
     data: KikuFieldGroupingRequestData,
   ) => Promise<KikuFieldGroupingChoice>;
   getKnownWordCacheStatePath: () => string;
+  getCachedMediaPath?: (
+    currentVideoPath: string,
+    kind: 'audio' | 'video',
+  ) => Promise<string | null>;
+  shouldRequireRemoteMediaCache?: () => boolean;
   shouldStartAnkiIntegration: () => boolean;
   bindOverlayOwner?: () => void;
   releaseOverlayOwner?: () => void;
@@ -97,6 +107,10 @@ export function createBuildInitializeOverlayRuntimeOptionsHandler(deps: {
     showOverlayNotification: deps.showOverlayNotification,
     createFieldGroupingCallback: deps.createFieldGroupingCallback,
     getKnownWordCacheStatePath: deps.getKnownWordCacheStatePath,
+    ...(deps.getCachedMediaPath ? { getCachedMediaPath: deps.getCachedMediaPath } : {}),
+    ...(deps.shouldRequireRemoteMediaCache
+      ? { shouldRequireRemoteMediaCache: deps.shouldRequireRemoteMediaCache }
+      : {}),
     shouldStartAnkiIntegration: deps.shouldStartAnkiIntegration,
     bindOverlayOwner: deps.bindOverlayOwner,
     releaseOverlayOwner: deps.releaseOverlayOwner,

src/main/runtime/youtube-playback-runtime.test.ts

@@ -146,3 +146,135 @@ test('youtube playback runtime resolves the socket path lazily for windows start
 
   assert.ok(calls.some((entry) => entry.includes('--input-ipc-server=/tmp/updated.sock')));
 });
+
+test('youtube playback runtime starts media cache without blocking the subtitle flow', async () => {
+  const calls: string[] = [];
+  let resolveCache: (() => void) | undefined;
+  const cachePromise = new Promise<void>((resolve) => {
+    resolveCache = resolve;
+  });
+
+  const runtime = createYoutubePlaybackRuntime({
+    platform: 'linux',
+    directPlaybackFormat: 'best',
+    mpvYtdlFormat: 'bestvideo+bestaudio',
+    autoLaunchTimeoutMs: 2_000,
+    connectTimeoutMs: 1_000,
+    getSocketPath: () => '/tmp/mpv.sock',
+    getMpvConnected: () => true,
+    invalidatePendingAutoplayReadyFallbacks: () => {
+      calls.push('invalidate-autoplay');
+    },
+    setAppOwnedFlowInFlight: (next) => {
+      calls.push(`app-owned:${next}`);
+    },
+    ensureYoutubePlaybackRuntimeReady: async () => {
+      calls.push('ensure-runtime-ready');
+    },
+    resolveYoutubePlaybackUrl: async () => {
+      throw new Error('linux path should not resolve direct playback url');
+    },
+    launchWindowsMpv: async () => ({ ok: false }),
+    waitForYoutubeMpvConnected: async () => true,
+    prepareYoutubePlaybackInMpv: async ({ url }) => {
+      calls.push(`prepare:${url}`);
+      return true;
+    },
+    startYoutubeMediaCache: async (url) => {
+      calls.push(`cache:${url}`);
+      await cachePromise;
+      calls.push('cache-done');
+    },
+    runYoutubePlaybackFlow: async ({ url, mode }) => {
+      calls.push(`run-flow:${url}:${mode}`);
+    },
+    logInfo: (message) => {
+      calls.push(`info:${message}`);
+    },
+    logWarn: (message) => {
+      calls.push(`warn:${message}`);
+    },
+    schedule: () => 1 as never,
+    clearScheduled: () => {},
+  });
+
+  await runtime.runYoutubePlaybackFlow({
+    url: 'https://youtu.be/demo',
+    mode: 'download',
+    source: 'second-instance',
+  });
+
+  const prepareIndex = calls.indexOf('prepare:https://youtu.be/demo');
+  const cacheIndex = calls.indexOf('cache:https://youtu.be/demo');
+  const runFlowIndex = calls.indexOf('run-flow:https://youtu.be/demo:download');
+  assert.notEqual(prepareIndex, -1);
+  assert.notEqual(cacheIndex, -1);
+  assert.notEqual(runFlowIndex, -1);
+  assert.ok(prepareIndex < cacheIndex);
+  assert.ok(cacheIndex < runFlowIndex);
+  assert.equal(calls.includes('cache-done'), false);
+  const resolveCacheNow = resolveCache;
+  assert.ok(resolveCacheNow);
+  resolveCacheNow();
+});
+
+test('youtube playback runtime logs synchronous media cache startup failures', async () => {
+  const calls: string[] = [];
+
+  const runtime = createYoutubePlaybackRuntime({
+    platform: 'linux',
+    directPlaybackFormat: 'best',
+    mpvYtdlFormat: 'bestvideo+bestaudio',
+    autoLaunchTimeoutMs: 2_000,
+    connectTimeoutMs: 1_000,
+    getSocketPath: () => '/tmp/mpv.sock',
+    getMpvConnected: () => true,
+    invalidatePendingAutoplayReadyFallbacks: () => {
+      calls.push('invalidate-autoplay');
+    },
+    setAppOwnedFlowInFlight: (next) => {
+      calls.push(`app-owned:${next}`);
+    },
+    ensureYoutubePlaybackRuntimeReady: async () => {
+      calls.push('ensure-runtime-ready');
+    },
+    resolveYoutubePlaybackUrl: async () => {
+      throw new Error('linux path should not resolve direct playback url');
+    },
+    launchWindowsMpv: async () => ({ ok: false }),
+    waitForYoutubeMpvConnected: async () => true,
+    prepareYoutubePlaybackInMpv: async ({ url }) => {
+      calls.push(`prepare:${url}`);
+      return true;
+    },
+    startYoutubeMediaCache: () => {
+      calls.push('cache');
+      throw new Error('cache exploded');
+    },
+    runYoutubePlaybackFlow: async ({ url, mode }) => {
+      calls.push(`run-flow:${url}:${mode}`);
+    },
+    logInfo: (message) => {
+      calls.push(`info:${message}`);
+    },
+    logWarn: (message) => {
+      calls.push(`warn:${message}`);
+    },
+    schedule: () => 1 as never,
+    clearScheduled: () => {},
+  });
+
+  await runtime.runYoutubePlaybackFlow({
+    url: 'https://youtu.be/demo',
+    mode: 'download',
+    source: 'second-instance',
+  });
+  await Promise.resolve();
+
+  assert.ok(calls.includes('run-flow:https://youtu.be/demo:download'));
+  assert.ok(
+    calls.some((entry) =>
+      entry.startsWith('warn:Failed to start YouTube media cache: cache exploded'),
+    ),
+  );
+});

src/main/runtime/youtube-playback-runtime.ts

@@ -20,6 +20,7 @@ export type YoutubePlaybackRuntimeDeps = {
   launchWindowsMpv: (playbackUrl: string, args: string[]) => Promise<LaunchResult>;
   waitForYoutubeMpvConnected: (timeoutMs: number) => Promise<boolean>;
   prepareYoutubePlaybackInMpv: (request: { url: string }) => Promise<boolean>;
+  startYoutubeMediaCache?: (url: string) => void | Promise<void>;
   runYoutubePlaybackFlow: (request: {
     url: string;
     mode: NonNullable<CliArgs['youtubeMode']>;
@@ -126,6 +127,18 @@ export function createYoutubePlaybackRuntime(deps: YoutubePlaybackRuntimeDeps) {
       if (!mediaReady) {
         throw new Error('Timed out waiting for mpv to load the requested YouTube URL.');
       }
+      if (deps.startYoutubeMediaCache) {
+        void new Promise<void>((resolve) => {
+          resolve(deps.startYoutubeMediaCache?.(request.url));
+        })
+          .catch((error) => {
+            deps.logWarn(
+              `Failed to start YouTube media cache: ${
+                error instanceof Error ? error.message : String(error)
+              }`,
+            );
+          });
+      }
 
       await deps.runYoutubePlaybackFlow({
         url: request.url,

src/media-generator.test.ts

@@ -8,6 +8,10 @@ import { buildAnimatedImageVideoFilter, MediaGenerator } from './media-generator
 
 async function withStubbedFfmpeg(
   run: (generator: MediaGenerator, argsPath: string) => Promise<void>,
+  options: {
+    logDebug?: (message: string) => void;
+    now?: () => number;
+  } = {},
 ): Promise<void> {
   const root = fs.mkdtempSync(path.join(os.tmpdir(), 'subminer-media-generator-test-'));
   const binDir = path.join(root, 'bin');
@@ -25,7 +29,7 @@ async function withStubbedFfmpeg(
       "  console.log(' V..... libaom-av1');",
       '  process.exit(0);',
       '}',
-      "fs.writeFileSync(process.env.SUBMINER_TEST_FFMPEG_ARGS, `${args.join('\\n')}\\n`, 'utf8');",
+      "fs.writeFileSync(process.env.SUBMINER_TEST_FFMPEG_ARGS, JSON.stringify(args), 'utf8');",
       'const outputPath = args.at(-1);',
       "fs.writeFileSync(outputPath, 'avif', 'utf8');",
     ].join('\n'),
@@ -44,7 +48,7 @@ async function withStubbedFfmpeg(
   const originalArgsPath = process.env.SUBMINER_TEST_FFMPEG_ARGS;
   process.env.PATH = `${binDir}${path.delimiter}${originalPath ?? ''}`;
   process.env.SUBMINER_TEST_FFMPEG_ARGS = argsPath;
-  const generator = new MediaGenerator(tempDir);
+  const generator = new MediaGenerator(tempDir, options);
 
   try {
     await run(generator, argsPath);
@@ -61,7 +65,7 @@ async function withStubbedFfmpeg(
 }
 
 function readFfmpegArgs(argsPath: string): string[] {
-  return fs.readFileSync(argsPath, 'utf8').trim().split('\n');
+  return JSON.parse(fs.readFileSync(argsPath, 'utf8')) as string[];
 }
 
 test('buildAnimatedImageVideoFilter holds lead-in until the next frame after the audio boundary', () => {
@@ -182,3 +186,128 @@ test('generateAudio recreates missing temp directory before invoking ffmpeg', as
     assert.equal(fs.existsSync(path.dirname(outputPath!)), true);
   });
 });
+
+test('generateAudio adds remote input options before the ffmpeg input', async () => {
+  await withStubbedFfmpeg(async (generator, argsPath) => {
+    await generator.generateAudio(
+      {
+        path: 'https://rr1---sn.example.googlevideo.com/videoplayback?mime=audio%2Fwebm',
+        inputOptions: {
+          reconnect: true,
+          userAgent: 'Mozilla/5.0',
+          headers: {
+            Referer: 'https://www.youtube.com/',
+            Origin: 'https://www.youtube.com',
+          },
+        },
+      },
+      10,
+      12,
+    );
+
+    const args = readFfmpegArgs(argsPath);
+    const inputIndex = args.indexOf('-i');
+    assert.ok(inputIndex > 0);
+    assert.ok(args.indexOf('-reconnect') > -1);
+    assert.ok(args.indexOf('-reconnect') < inputIndex);
+    assert.equal(args[args.indexOf('-reconnect') + 1], '1');
+    assert.equal(args[args.indexOf('-reconnect_streamed') + 1], '1');
+    assert.equal(args[args.indexOf('-reconnect_on_network_error') + 1], '1');
+    assert.equal(args[args.indexOf('-reconnect_on_http_error') + 1], '403,5xx');
+    assert.equal(args[args.indexOf('-reconnect_delay_max') + 1], '5');
+    assert.equal(args[args.indexOf('-user_agent') + 1], 'Mozilla/5.0');
+    assert.equal(
+      args[args.indexOf('-headers') + 1],
+      'Referer: https://www.youtube.com/\r\nOrigin: https://www.youtube.com\r\n',
+    );
+  });
+});
+
+test('generateAudio skips stale audio stream maps for single resolved streams', async () => {
+  await withStubbedFfmpeg(async (generator, argsPath) => {
+    await generator.generateAudio(
+      {
+        path: 'https://rr1---sn.example.googlevideo.com/videoplayback?mime=audio%2Fwebm',
+        singleResolvedStream: true,
+      },
+      10,
+      12,
+      0,
+      22,
+    );
+
+    const args = readFfmpegArgs(argsPath);
+    assert.equal(args.includes('-map'), false);
+  });
+});
+
+test('generateAudio keeps explicit audio stream maps for normal media paths', async () => {
+  await withStubbedFfmpeg(async (generator, argsPath) => {
+    await generator.generateAudio('/video.mp4', 10, 12, 0, 2);
+
+    const args = readFfmpegArgs(argsPath);
+    assert.equal(args[args.indexOf('-map') + 1], '0:2');
+  });
+});
+
+test('generateAudio debug-logs cached input and completion timing', async () => {
+  const logs: string[] = [];
+  const times = [1000, 1052];
+
+  await withStubbedFfmpeg(
+    async (generator) => {
+      await generator.generateAudio(
+        {
+          path: '/tmp/subminer-youtube-media-cache/abc123/media.mkv',
+          source: 'youtube-cache',
+        },
+        10,
+        12,
+      );
+    },
+    {
+      logDebug: (message) => logs.push(message),
+      now: () => times.shift() ?? 1052,
+    },
+  );
+
+  assert.match(logs.join('\n'), /\[media-generator\] audio start/);
+  assert.match(logs.join('\n'), /source=youtube-cache/);
+  assert.match(
+    logs.join('\n'),
+    /input=local:\/tmp\/subminer-youtube-media-cache\/abc123\/media\.mkv/,
+  );
+  assert.match(logs.join('\n'), /\[media-generator\] audio complete/);
+  assert.match(logs.join('\n'), /elapsedMs=52/);
+  assert.match(logs.join('\n'), /bytes=4/);
+});
+
+test('generateAudio debug logs sanitize remote inputs', async () => {
+  const logs: string[] = [];
+  const times = [1000, 1003];
+
+  await withStubbedFfmpeg(
+    async (generator) => {
+      await generator.generateAudio(
+        {
+          path: 'https://rr1---sn.example.googlevideo.com/videoplayback?signature=secret&expire=123',
+          inputOptions: {
+            reconnect: true,
+            headers: {
+              Referer: 'https://www.youtube.com/watch?v=abc123',
+            },
+          },
+        },
+        10,
+        12,
+      );
+    },
+    {
+      logDebug: (message) => logs.push(message),
+      now: () => times.shift() ?? 1003,
+    },
+  );
+
+  assert.match(logs.join('\n'), /input=remote:rr1---sn\.example\.googlevideo\.com/);
+  assert.doesNotMatch(logs.join('\n'), /signature=secret|expire=123|Referer|abc123/);
+});

src/media-generator.ts

@@ -21,9 +21,12 @@ import * as fs from 'fs';
 import * as path from 'path';
 import * as os from 'os';
 import { createLogger } from './logger';
+import { normalizeMediaInput, type MediaInput } from './media-input';
 
 const log = createLogger('media');
 
+export type { MediaInput, MediaInputOptions } from './media-input';
+
 function normalizeAnimatedImageFps(fps: number | undefined): number {
   const fallbackFps = 10;
   const safeFps = typeof fps === 'number' && Number.isFinite(fps) ? fps : fallbackFps;
@@ -69,12 +72,65 @@ export function buildAnimatedImageVideoFilter(options: {
   return vfParts.join(',');
 }
 
+export interface MediaGeneratorOptions {
+  logDebug?: (message: string) => void;
+  now?: () => number;
+}
+
+function sanitizeDebugToken(value: string, fallback: string): string {
+  const trimmed = value.trim();
+  if (!trimmed) {
+    return fallback;
+  }
+  const sanitized = trimmed.replace(/[^A-Za-z0-9_.:-]+/g, '-').slice(0, 80);
+  return sanitized || fallback;
+}
+
+function describeMediaInputPathForDebugLog(value: string): string {
+  try {
+    const url = new URL(value);
+    if (url.protocol === 'http:' || url.protocol === 'https:') {
+      return `remote:${url.hostname.toLowerCase() || 'unknown'}`;
+    }
+    return `${url.protocol.replace(/:$/, '')}:`;
+  } catch {
+    // Not a URL; treat as a local file path below.
+  }
+
+  if (value.startsWith('edl://')) {
+    return 'edl:';
+  }
+
+  return `local:${value}`;
+}
+
+function describeMediaInputForDebugLog(input: MediaInput): string {
+  const pathValue = typeof input === 'string' ? input : input.path;
+  const sourceValue = typeof input === 'string' ? 'raw' : input.source;
+  const source = sanitizeDebugToken(sourceValue ?? 'raw', 'raw');
+  return `source=${source} input=${describeMediaInputPathForDebugLog(pathValue)}`;
+}
+
+function describeFfmpegFailureForDebugLog(error: ExecFileException): string {
+  const code = typeof error.code === 'string' || typeof error.code === 'number' ? error.code : null;
+  const signal = typeof error.signal === 'string' ? error.signal : null;
+  if (code !== null) {
+    return `code=${code}`;
+  }
+  if (signal) {
+    return `signal=${signal}`;
+  }
+  return `name=${sanitizeDebugToken(error.name || 'Error', 'Error')}`;
+}
+
 export class MediaGenerator {
   private tempDir: string;
   private notifyIconDir: string;
   private av1EncoderPromise: Promise<string | null> | null = null;
+  private readonly options: MediaGeneratorOptions;
 
-  constructor(tempDir?: string) {
+  constructor(tempDir?: string, options: MediaGeneratorOptions = {}) {
+    this.options = options;
     this.tempDir = tempDir || path.join(os.tmpdir(), 'subminer-media');
     this.notifyIconDir = path.join(os.tmpdir(), 'subminer-notify');
     this.ensureDirectory(this.tempDir);
@@ -83,6 +139,28 @@ export class MediaGenerator {
     this.cleanupOldNotificationIcons();
   }
 
+  private nowMs(): number {
+    try {
+      const value = this.options.now?.() ?? Date.now();
+      return Number.isFinite(value) ? value : Date.now();
+    } catch {
+      return Date.now();
+    }
+  }
+
+  private elapsedMs(startedAt: number): number {
+    return Math.max(0, Math.round(this.nowMs() - startedAt));
+  }
+
+  private logMediaDebug(message: string): void {
+    const logDebug = this.options.logDebug ?? ((line: string) => log.debug(line));
+    try {
+      logDebug(`[media-generator] ${message}`);
+    } catch {
+      // Debug logging should not affect media generation.
+    }
+  }
+
   private ensureDirectory(dir: string): void {
     if (!fs.existsSync(dir)) {
       fs.mkdirSync(dir, { recursive: true });
@@ -181,7 +259,7 @@ export class MediaGenerator {
   }
 
   async generateAudio(
-    videoPath: string,
+    videoPath: MediaInput,
     startTime: number,
     endTime: number,
     padding: number = 0,
@@ -190,12 +268,24 @@ export class MediaGenerator {
     const safePadding = Number.isFinite(padding) ? Math.max(0, padding) : 0;
     const start = Math.max(0, startTime - safePadding);
     const duration = endTime - start + safePadding;
+    const mediaInput = normalizeMediaInput(videoPath);
+    const inputDescription = describeMediaInputForDebugLog(videoPath);
 
     return new Promise((resolve, reject) => {
       const outputPath = this.createTempOutputPath('audio', 'mp3');
-      const args: string[] = ['-ss', start.toString(), '-t', duration.toString(), '-i', videoPath];
+      const startedAt = this.nowMs();
+      const args: string[] = [
+        '-ss',
+        start.toString(),
+        '-t',
+        duration.toString(),
+        ...mediaInput.inputArgs,
+        '-i',
+        mediaInput.path,
+      ];
 
       if (
+        !mediaInput.singleResolvedStream &&
         typeof audioStreamIndex === 'number' &&
         Number.isInteger(audioStreamIndex) &&
         audioStreamIndex >= 0
@@ -205,8 +295,14 @@ export class MediaGenerator {
 
       args.push('-vn', '-acodec', 'libmp3lame', '-q:a', '2', '-ar', '44100', '-y', outputPath);
 
+      this.logMediaDebug(
+        `audio start ${inputDescription} start=${start} duration=${duration} padding=${safePadding}`,
+      );
       execFile('ffmpeg', args, { timeout: 30000 }, (error) => {
         if (error) {
+          this.logMediaDebug(
+            `audio failed ${inputDescription} elapsedMs=${this.elapsedMs(startedAt)} ${describeFfmpegFailureForDebugLog(error)}`,
+          );
           reject(this.ffmpegError('audio generation', error));
           return;
         }
@@ -214,6 +310,9 @@ export class MediaGenerator {
         try {
           const data = fs.readFileSync(outputPath);
           fs.unlinkSync(outputPath);
+          this.logMediaDebug(
+            `audio complete ${inputDescription} elapsedMs=${this.elapsedMs(startedAt)} bytes=${data.byteLength}`,
+          );
           resolve(data);
         } catch (err) {
           reject(err);
@@ -223,7 +322,7 @@ export class MediaGenerator {
   }
 
   async generateScreenshot(
-    videoPath: string,
+    videoPath: MediaInput,
     timestamp: number,
     options: {
       format: 'jpg' | 'png' | 'webp';
@@ -239,8 +338,18 @@ export class MediaGenerator {
       png: 'png',
       webp: 'webp',
     };
+    const mediaInput = normalizeMediaInput(videoPath);
+    const inputDescription = describeMediaInputForDebugLog(videoPath);
 
-    const args: string[] = ['-ss', timestamp.toString(), '-i', videoPath, '-vframes', '1'];
+    const args: string[] = [
+      '-ss',
+      timestamp.toString(),
+      ...mediaInput.inputArgs,
+      '-i',
+      mediaInput.path,
+      '-vframes',
+      '1',
+    ];
 
     const vfParts: string[] = [];
     if (maxWidth && maxWidth > 0 && maxHeight && maxHeight > 0) {
@@ -270,10 +379,17 @@ export class MediaGenerator {
 
     return new Promise((resolve, reject) => {
       const outputPath = this.createTempOutputPath('screenshot', ext);
+      const startedAt = this.nowMs();
       args.push(outputPath);
 
+      this.logMediaDebug(
+        `screenshot start ${inputDescription} timestamp=${timestamp} format=${format} maxWidth=${maxWidth ?? 'none'} maxHeight=${maxHeight ?? 'none'}`,
+      );
       execFile('ffmpeg', args, { timeout: 30000 }, (error) => {
         if (error) {
+          this.logMediaDebug(
+            `screenshot failed ${inputDescription} elapsedMs=${this.elapsedMs(startedAt)} ${describeFfmpegFailureForDebugLog(error)}`,
+          );
           reject(this.ffmpegError('screenshot generation', error));
           return;
         }
@@ -281,6 +397,9 @@ export class MediaGenerator {
         try {
           const data = fs.readFileSync(outputPath);
           fs.unlinkSync(outputPath);
+          this.logMediaDebug(
+            `screenshot complete ${inputDescription} elapsedMs=${this.elapsedMs(startedAt)} bytes=${data.byteLength}`,
+          );
           resolve(data);
         } catch (err) {
           reject(err);
@@ -334,7 +453,7 @@ export class MediaGenerator {
   }
 
   async generateAnimatedImage(
-    videoPath: string,
+    videoPath: MediaInput,
     startTime: number,
     endTime: number,
     padding: number = 0,
@@ -352,10 +471,15 @@ export class MediaGenerator {
     const start = Math.max(0, startTime - safePadding);
     const duration = roundDurationUpToNextFrameBoundary(endTime - start + safePadding, clampedFps);
     const totalLeadingStillDuration = Math.max(0, leadingStillDuration);
+    const inputDescription = describeMediaInputForDebugLog(videoPath);
 
     const clampedCrf = Math.max(0, Math.min(63, crf));
 
+    const encoderDetectionStartedAt = this.nowMs();
     const av1Encoder = await this.detectAv1Encoder();
+    this.logMediaDebug(
+      `animated-image encoder ${inputDescription} elapsedMs=${this.elapsedMs(encoderDetectionStartedAt)} encoder=${av1Encoder ?? 'none'}`,
+    );
     if (!av1Encoder) {
       throw new Error(
         'No supported AV1 encoder found for animated AVIF (tried libaom-av1, libsvtav1, librav1e).',
@@ -364,6 +488,8 @@ export class MediaGenerator {
 
     return new Promise((resolve, reject) => {
       const outputPath = this.createTempOutputPath('animation', 'avif');
+      const mediaInput = normalizeMediaInput(videoPath);
+      const startedAt = this.nowMs();
 
       const encoderArgs: string[] = ['-c:v', av1Encoder];
       if (av1Encoder === 'libaom-av1') {
@@ -375,6 +501,9 @@ export class MediaGenerator {
         encoderArgs.push('-qp', clampedCrf.toString(), '-speed', '8');
       }
 
+      this.logMediaDebug(
+        `animated-image start ${inputDescription} start=${start} duration=${duration} padding=${safePadding} fps=${clampedFps} maxWidth=${maxWidth ?? 'none'} maxHeight=${maxHeight ?? 'none'} crf=${clampedCrf} encoder=${av1Encoder}`,
+      );
       execFile(
         'ffmpeg',
         [
@@ -382,8 +511,9 @@ export class MediaGenerator {
           start.toString(),
           '-t',
           duration.toString(),
+          ...mediaInput.inputArgs,
           '-i',
-          videoPath,
+          mediaInput.path,
           '-vf',
           buildAnimatedImageVideoFilter({
             fps: clampedFps,
@@ -398,6 +528,9 @@ export class MediaGenerator {
         { timeout: 60000 },
         (error) => {
           if (error) {
+            this.logMediaDebug(
+              `animated-image failed ${inputDescription} elapsedMs=${this.elapsedMs(startedAt)} ${describeFfmpegFailureForDebugLog(error)}`,
+            );
             reject(this.ffmpegError('animation generation', error));
             return;
           }
@@ -405,6 +538,9 @@ export class MediaGenerator {
           try {
             const data = fs.readFileSync(outputPath);
             fs.unlinkSync(outputPath);
+            this.logMediaDebug(
+              `animated-image complete ${inputDescription} elapsedMs=${this.elapsedMs(startedAt)} bytes=${data.byteLength}`,
+            );
             resolve(data);
           } catch (err) {
             reject(err);

src/media-input.ts

@@ -0,0 +1,97 @@
+export interface MediaInputOptions {
+  reconnect?: boolean;
+  userAgent?: string;
+  headers?: Record<string, string>;
+}
+
+export type MediaInput =
+  | string
+  | {
+      path: string;
+      source?: string;
+      inputOptions?: MediaInputOptions;
+      singleResolvedStream?: boolean;
+    };
+
+export type NormalizedMediaInput = {
+  path: string;
+  inputArgs: string[];
+  singleResolvedStream: boolean;
+};
+
+const BLOCKED_FFMPEG_HEADER_NAMES = new Set(['authorization', 'cookie', 'proxy-authorization']);
+
+function trimToNonEmptyString(value: unknown): string | null {
+  if (typeof value !== 'string') {
+    return null;
+  }
+  const trimmed = value.trim();
+  return trimmed.length > 0 ? trimmed : null;
+}
+
+function normalizeHeaderName(value: string): string | null {
+  const trimmed = value.trim();
+  if (!/^[A-Za-z0-9!#$%&'*+.^_`|~-]+$/.test(trimmed)) {
+    return null;
+  }
+  if (BLOCKED_FFMPEG_HEADER_NAMES.has(trimmed.toLowerCase())) {
+    return null;
+  }
+  return trimmed;
+}
+
+function serializeFfmpegHeaders(headers: Record<string, string> | undefined): string | null {
+  if (!headers) {
+    return null;
+  }
+
+  const lines: string[] = [];
+  for (const [rawName, rawValue] of Object.entries(headers)) {
+    const name = normalizeHeaderName(rawName);
+    const value = trimToNonEmptyString(rawValue);
+    if (!name || !value) {
+      continue;
+    }
+    lines.push(`${name}: ${value.replace(/[\r\n]+/g, ' ')}`);
+  }
+
+  return lines.length > 0 ? `${lines.join('\r\n')}\r\n` : null;
+}
+
+export function normalizeMediaInput(input: MediaInput): NormalizedMediaInput {
+  if (typeof input === 'string') {
+    return { path: input, inputArgs: [], singleResolvedStream: false };
+  }
+
+  const inputArgs: string[] = [];
+  if (input.inputOptions?.reconnect) {
+    inputArgs.push(
+      '-reconnect',
+      '1',
+      '-reconnect_streamed',
+      '1',
+      '-reconnect_on_network_error',
+      '1',
+      '-reconnect_on_http_error',
+      '403,5xx',
+      '-reconnect_delay_max',
+      '5',
+    );
+  }
+
+  const userAgent = trimToNonEmptyString(input.inputOptions?.userAgent);
+  if (userAgent) {
+    inputArgs.push('-user_agent', userAgent);
+  }
+
+  const headers = serializeFfmpegHeaders(input.inputOptions?.headers);
+  if (headers) {
+    inputArgs.push('-headers', headers);
+  }
+
+  return {
+    path: input.path,
+    inputArgs,
+    singleResolvedStream: input.singleResolvedStream === true,
+  };
+}

src/settings/settings-controls.test.ts

@@ -93,3 +93,32 @@ test('select controls show config-only current values without offering them othe
     restoreDocument();
   }
 });
+
+test('select controls use configured option labels', () => {
+  const restoreDocument = installDocumentStub();
+  try {
+    const field = {
+      ...createSelectField(),
+      enumValues: ['direct', 'background'],
+      enumLabels: {
+        direct: 'Direct stream extraction',
+        background: 'Background media cache',
+      },
+    } as ConfigSettingsField & { enumLabels: Record<string, string> };
+
+    const control = renderControl(field, {
+      valueForField: () => 'direct',
+      valueForPath: () => undefined,
+      updateDraft: () => {},
+      resetDraftPath: () => {},
+      setFieldError: () => {},
+    }) as HTMLSelectElement;
+
+    assert.deepEqual(
+      Array.from(control.options).map((option) => option.textContent),
+      ['Direct stream extraction', 'Background media cache'],
+    );
+  } finally {
+    restoreDocument();
+  }
+});

src/settings/settings-controls.ts

@@ -227,7 +227,7 @@ export function renderControl(
     for (const enumValue of enumValues) {
       const option = createElement('option') as HTMLOptionElement;
       option.value = enumValue;
-      option.textContent = enumValue;
+      option.textContent = field.enumLabels?.[enumValue] ?? enumValue;
       option.selected = enumValue === value;
       select.append(option);
     }

src/types/config.ts

@@ -16,6 +16,7 @@ import type {
   StatsConfig,
   YomitanConfig,
   YoutubeConfig,
+  YoutubeMediaCacheMode,
   YoutubeSubgenConfig,
 } from './integrations';
 import type {
@@ -345,6 +346,10 @@ export interface ResolvedConfig {
   };
   youtube: YoutubeConfig & {
     primarySubLanguages: string[];
+    mediaCache: {
+      mode: YoutubeMediaCacheMode;
+      maxHeight: number;
+    };
   };
   youtubeSubgen: YoutubeSubgenConfig & {
     whisperBin: string;

src/types/integrations.ts

@@ -122,8 +122,16 @@ export interface AiConfig {
   requestTimeoutMs?: number;
 }
 
+export type YoutubeMediaCacheMode = 'direct' | 'background';
+
+export interface YoutubeMediaCacheConfig {
+  mode?: YoutubeMediaCacheMode;
+  maxHeight?: number;
+}
+
 export interface YoutubeConfig {
   primarySubLanguages?: string[];
+  mediaCache?: YoutubeMediaCacheConfig;
 }
 
 export interface YoutubeSubgenConfig {

src/types/settings.ts

@@ -43,6 +43,7 @@ export interface ConfigSettingsField {
   control: ConfigSettingsControl;
   defaultValue: unknown;
   enumValues?: readonly string[];
+  enumLabels?: Record<string, string>;
   restartBehavior: ConfigSettingsRestartBehavior;
   advanced?: boolean;
   secret?: boolean;