@steipete/summarize 0.11.0 → 0.12.0

package/dist/types/slides/download.d.ts

@@ -0,0 +1,29 @@
+export declare function buildYtDlpCookiesArgs(cookiesFromBrowser?: string | null): string[];
+export declare function buildSlidesMediaCacheKey(url: string): string;
+export declare function formatBytes(bytes: number): string;
+export declare function downloadYoutubeVideo(options: {
+    ytDlpPath: string;
+    url: string;
+    timeoutMs: number;
+    format: string;
+    cookiesFromBrowser?: string | null;
+    onProgress?: ((percent: number, detail?: string) => void) | null;
+}): Promise<{
+    filePath: string;
+    cleanup: () => Promise<void>;
+}>;
+export declare function downloadRemoteVideo(options: {
+    url: string;
+    timeoutMs: number;
+    onProgress?: ((percent: number, detail?: string) => void) | null;
+}): Promise<{
+    filePath: string;
+    cleanup: () => Promise<void>;
+}>;
+export declare function resolveYoutubeStreamUrl(options: {
+    ytDlpPath: string;
+    url: string;
+    timeoutMs: number;
+    format: string;
+    cookiesFromBrowser?: string | null;
+}): Promise<string>;

package/dist/types/slides/extract-finalize.d.ts

@@ -0,0 +1,57 @@
+import type { SlideAutoTune, SlideExtractionResult, SlideImage, SlideSource, SlideSourceKind } from "./types.js";
+export declare const SLIDES_PROGRESS: {
+    readonly PREPARE: 2;
+    readonly FETCH_VIDEO: 6;
+    readonly DOWNLOAD_VIDEO: 35;
+    readonly DETECT_SCENES: 60;
+    readonly EXTRACT_FRAMES: 90;
+    readonly OCR: 99;
+    readonly FINAL: 100;
+};
+export type SlidesChunkMeta = {
+    slidesDir: string;
+    sourceUrl: string;
+    sourceId: string;
+    sourceKind: SlideSourceKind;
+    ocrAvailable: boolean;
+};
+export declare function buildSlidesChunkMeta(args: {
+    slidesDir: string;
+    source: SlideSource;
+    ocrAvailable: boolean;
+}): SlidesChunkMeta;
+export declare function buildSlideTimeline(args: {
+    source: SlideSource;
+    slidesDir: string;
+    sceneThreshold: number;
+    autoTuneThreshold: boolean;
+    autoTune: SlideAutoTune;
+    maxSlides: number;
+    minSlideDuration: number;
+    ocrRequested: boolean;
+    ocrAvailable: boolean;
+    warnings: string[];
+    slides: Array<SlideImage & {
+        segment?: unknown;
+    }>;
+}): SlideExtractionResult;
+export declare function emitPlaceholderSlides(args: {
+    slides: Array<SlideImage & {
+        segment?: unknown;
+    }>;
+    meta: SlidesChunkMeta;
+    onSlideChunk?: ((value: {
+        slide: SlideImage;
+        meta: SlidesChunkMeta;
+    }) => void) | null;
+}): void;
+export declare function emitFinalSlides(args: {
+    slides: SlideImage[];
+    meta: SlidesChunkMeta;
+    onSlideChunk?: ((value: {
+        slide: SlideImage;
+        meta: SlidesChunkMeta;
+    }) => void) | null;
+}): void;
+export declare function renameSlidesWithTimestamps(slides: SlideImage[], slidesDir: string): Promise<SlideImage[]>;
+export declare function writeSlidesJson(result: SlideExtractionResult, slidesDir: string): Promise<void>;

package/dist/types/slides/extract.d.ts

@@ -1,6 +1,7 @@
 import type { ExtractedLinkContent, MediaCache } from "../content/index.js";
 import type { SlideSettings } from "./settings.js";
 import type { SlideExtractionResult, SlideImage, SlideSource, SlideSourceKind } from "./types.js";
+export { parseShowinfoTimestamp, resolveExtractedTimestamp } from "./scene-detection.js";
 type ExtractSlidesArgs = {
     source: SlideSource;
     settings: SlideSettings;
@@ -34,10 +35,3 @@ export declare function resolveSlideSource({ url, extracted, }: {
 }): SlideSource | null;
 export declare function resolveSlideSourceFromUrl(url: string): SlideSource | null;
 export declare function extractSlidesForSource({ source, settings, noCache, mediaCache, env, timeoutMs, ytDlpPath, ytDlpCookiesFromBrowser, ffmpegPath, tesseractPath, hooks, }: ExtractSlidesArgs): Promise<SlideExtractionResult>;
-export declare function parseShowinfoTimestamp(line: string): number | null;
-export declare function resolveExtractedTimestamp({ requested, actual, seekBase, }: {
-    requested: number;
-    actual: number | null;
-    seekBase?: number | null;
-}): number;
-export {};

package/dist/types/slides/frame-extraction.d.ts

@@ -0,0 +1,38 @@
+import type { SlideAutoTune, SlideImage } from "./types.js";
+export declare function detectSlideTimestamps({ ffmpegPath, ffprobePath, inputPath, sceneThreshold, autoTuneThreshold, env, timeoutMs, warnings, workers, sampleCount, onSegmentProgress, logSlides, logSlidesTiming, }: {
+    ffmpegPath: string;
+    ffprobePath: string | null;
+    inputPath: string;
+    sceneThreshold: number;
+    autoTuneThreshold: boolean;
+    env: Record<string, string | undefined>;
+    timeoutMs: number;
+    warnings: string[];
+    workers: number;
+    sampleCount: number;
+    onSegmentProgress?: ((completed: number, total: number) => void) | null;
+    logSlides?: ((message: string) => void) | null;
+    logSlidesTiming?: ((label: string, startedAt: number) => number) | null;
+}): Promise<{
+    timestamps: number[];
+    autoTune: SlideAutoTune;
+    durationSeconds: number | null;
+}>;
+export declare function extractFramesAtTimestamps({ ffmpegPath, inputPath, outputDir, timestamps, segments, durationSeconds, timeoutMs, workers, onProgress, onStatus, onSlide, logSlides, logSlidesTiming, }: {
+    ffmpegPath: string;
+    inputPath: string;
+    outputDir: string;
+    timestamps: number[];
+    segments?: Array<{
+        start: number;
+        end: number | null;
+    } | null>;
+    durationSeconds?: number | null;
+    timeoutMs: number;
+    workers: number;
+    onProgress?: ((completed: number, total: number) => void) | null;
+    onStatus?: ((text: string) => void) | null;
+    onSlide?: ((slide: SlideImage) => void) | null;
+    logSlides?: ((message: string) => void) | null;
+    logSlidesTiming?: ((label: string, startedAt: number) => number) | null;
+}): Promise<SlideImage[]>;

package/dist/types/slides/ingest.d.ts

@@ -0,0 +1,47 @@
+import type { MediaCache } from "../content/index.js";
+import type { SlideSource } from "./types.js";
+export type SlidesIngestProgress = (label: string, percent: number, detail?: string) => void;
+export declare function prepareSlidesInput({ source, mediaCache, timeoutMs, ytDlpPath, ytDlpCookiesFromBrowser, resolveSlidesYtDlpExtractFormat, resolveSlidesStreamFallback, buildSlidesMediaCacheKey, formatBytes, reportSlidesProgress, logSlidesTiming, downloadYoutubeVideo, downloadRemoteVideo, resolveYoutubeStreamUrl, }: {
+    source: SlideSource;
+    mediaCache: MediaCache | null;
+    timeoutMs: number;
+    ytDlpPath: string | null;
+    ytDlpCookiesFromBrowser?: string | null;
+    resolveSlidesYtDlpExtractFormat: () => string;
+    resolveSlidesStreamFallback: () => boolean;
+    buildSlidesMediaCacheKey: (url: string) => string;
+    formatBytes: (bytes: number) => string;
+    reportSlidesProgress?: SlidesIngestProgress | null;
+    logSlidesTiming?: ((label: string, startedAt: number) => number) | null;
+    downloadYoutubeVideo: (args: {
+        ytDlpPath: string;
+        url: string;
+        timeoutMs: number;
+        format: string;
+        cookiesFromBrowser?: string | null;
+        onProgress?: ((percent: number, detail?: string) => void) | null;
+    }) => Promise<{
+        filePath: string;
+        cleanup: () => Promise<void>;
+    }>;
+    downloadRemoteVideo: (args: {
+        url: string;
+        timeoutMs: number;
+        onProgress?: ((percent: number, detail?: string) => void) | null;
+    }) => Promise<{
+        filePath: string;
+        cleanup: () => Promise<void>;
+    }>;
+    resolveYoutubeStreamUrl: (args: {
+        ytDlpPath: string;
+        url: string;
+        timeoutMs: number;
+        format: string;
+        cookiesFromBrowser?: string | null;
+    }) => Promise<string>;
+}): Promise<{
+    inputPath: string;
+    inputCleanup: (() => Promise<void>) | null;
+    cachedMedia: Awaited<ReturnType<NonNullable<MediaCache>["get"]>> | null;
+    warnings: string[];
+}>;

package/dist/types/slides/ocr.d.ts

@@ -0,0 +1,5 @@
+import type { SlideImage } from "./types.js";
+export declare function cleanOcrText(text: string): string;
+export declare function estimateOcrConfidence(text: string): number;
+export declare function runTesseract(tesseractPath: string, imagePath: string): Promise<string>;
+export declare function runOcrOnSlides(slides: SlideImage[], tesseractPath: string, workers: number, onProgress?: ((completed: number, total: number) => void) | null): Promise<SlideImage[]>;

package/dist/types/slides/process.d.ts

@@ -0,0 +1,22 @@
+import type { ProcessHandle } from "../processes.js";
+export declare function runProcess({ command, args, timeoutMs, errorLabel, onStderrLine, onStdoutLine, }: {
+    command: string;
+    args: string[];
+    timeoutMs: number;
+    errorLabel: string;
+    onStderrLine?: (line: string, handle: ProcessHandle | null) => void;
+    onStdoutLine?: (line: string, handle: ProcessHandle | null) => void;
+}): Promise<void>;
+export declare function runProcessCapture({ command, args, timeoutMs, errorLabel, }: {
+    command: string;
+    args: string[];
+    timeoutMs: number;
+    errorLabel: string;
+}): Promise<string>;
+export declare function runProcessCaptureBuffer({ command, args, timeoutMs, errorLabel, }: {
+    command: string;
+    args: string[];
+    timeoutMs: number;
+    errorLabel: string;
+}): Promise<Buffer>;
+export declare function runWithConcurrency<T>(tasks: Array<() => Promise<T>>, workers: number, onProgress?: ((completed: number, total: number) => void) | null): Promise<T[]>;

package/dist/types/slides/scene-detection.d.ts

@@ -0,0 +1,75 @@
+import type { SlideImage } from "./types.js";
+export type SceneSegment = {
+    start: number;
+    end: number | null;
+};
+export declare function clamp(value: number, min: number, max: number): number;
+export declare function parseShowinfoTimestamp(line: string): number | null;
+export declare function resolveExtractedTimestamp({ requested, actual, seekBase, }: {
+    requested: number;
+    actual: number | null;
+    seekBase?: number | null;
+}): number;
+export declare function roundThreshold(value: number): number;
+export declare function calibrateSceneThreshold({ ffmpegPath, inputPath, durationSeconds, sampleCount, timeoutMs, logSlides, }: {
+    ffmpegPath: string;
+    inputPath: string;
+    durationSeconds: number | null;
+    sampleCount: number;
+    timeoutMs: number;
+    logSlides?: ((message: string) => void) | null;
+}): Promise<{
+    threshold: number;
+    confidence: number;
+}>;
+export declare function buildSegments(durationSeconds: number | null, workers: number): Array<{
+    start: number;
+    duration: number;
+}>;
+export declare function detectSceneTimestamps({ ffmpegPath, inputPath, threshold, timeoutMs, segments, workers, onSegmentProgress, runWithConcurrency, }: {
+    ffmpegPath: string;
+    inputPath: string;
+    threshold: number;
+    timeoutMs: number;
+    segments?: Array<{
+        start: number;
+        duration: number;
+    }>;
+    workers?: number;
+    onSegmentProgress?: ((completed: number, total: number) => void) | null;
+    runWithConcurrency: <T>(tasks: Array<() => Promise<T>>, workers: number, onProgress?: ((completed: number, total: number) => void) | null) => Promise<T[]>;
+}): Promise<number[]>;
+export declare function applyMinDurationFilter(slides: SlideImage[], minDurationSeconds: number, warnings: string[], removeFile: (path: string) => void): SlideImage[];
+export declare function mergeTimestamps(sceneTimestamps: number[], intervalTimestamps: number[], minDurationSeconds: number): number[];
+export declare function filterTimestampsByMinDuration(timestamps: number[], minDurationSeconds: number): number[];
+export declare function buildSceneSegments(sceneTimestamps: number[], durationSeconds: number | null): SceneSegment[];
+export declare function findSceneSegment(segments: SceneSegment[], timestamp: number): SceneSegment | null;
+export declare function adjustTimestampWithinSegment(timestamp: number, segment: SceneSegment | null): number;
+export declare function selectTimestampTargets({ targets, sceneTimestamps, minDurationSeconds, intervalSeconds, }: {
+    targets: number[];
+    sceneTimestamps: number[];
+    minDurationSeconds: number;
+    intervalSeconds: number;
+}): number[];
+export declare function buildIntervalTimestamps({ durationSeconds, minDurationSeconds, maxSlides, }: {
+    durationSeconds: number | null;
+    minDurationSeconds: number;
+    maxSlides: number;
+}): {
+    timestamps: number[];
+    intervalSeconds: number;
+} | null;
+export declare function probeVideoInfo({ ffprobePath, inputPath, timeoutMs, }: {
+    ffprobePath: string;
+    inputPath: string;
+    timeoutMs: number;
+}): Promise<{
+    durationSeconds: number | null;
+    width: number | null;
+    height: number | null;
+}>;
+export declare function applyMaxSlidesFilter<T extends {
+    index: number;
+    timestamp: number;
+    imagePath: string;
+}>(slides: T[], maxSlides: number, warnings: string[], removeFile: (path: string) => void): T[];

package/dist/types/slides/source-id.d.ts

@@ -0,0 +1,2 @@
+export declare function buildYoutubeSourceId(videoId: string): string;
+export declare function buildDirectSourceId(url: string): string;

package/dist/types/version.d.ts

@@ -1,4 +1,4 @@
-export declare const FALLBACK_VERSION = "0.11.0";
+export declare const FALLBACK_VERSION = "0.12.0";
 export declare function resolvePackageVersion(importMetaUrl?: string): string;
 export declare function resolveGitSha(importMetaUrl?: string): string | null;
 export declare function formatVersionLine(importMetaUrl?: string): string;

package/docs/_config.yml

@@ -1,5 +1,6 @@
 title: "summarize"
 description: "Summarize docs"
+baseurl: ""
 markdown: kramdown
 highlighter: rouge
 

package/docs/agent.md

@@ -135,10 +135,11 @@ Returns cached chat history for the same cache key as `/v1/agent`.
 ### Model Resolution (Daemon)
 
 - **Fixed model** (explicit `model`): parsed as `<provider>/<model>`. Provider base URL overrides come from config/env (OpenAI, Anthropic, Google, xAI, ZAI). OpenRouter uses OpenAI-compatible completions.
-- **Auto model**: uses existing auto-selection logic (`buildAutoModelAttempts`), skipping CLI transports.
+- **Auto model**: uses existing auto-selection logic (`buildAutoModelAttempts`), preferring API-key transports and then CLI fallback when available.
 - **Synthetic models**: created for OpenAI-compatible base URLs (local/openrouter).
 - `maxOutputTokens` defaults to 2048 or `maxOutputTokens` override.
-- CLI models are **not** supported in the daemon.
+- CLI models are supported as auto fallback and via explicit `cli/<provider>/<model>` overrides.
+- If the daemon still says no model is available after key/install changes, restart or reinstall it so the saved environment snapshot refreshes.
 
 ## Page Content Payload
 

package/docs/assets/site.css

@@ -184,6 +184,7 @@ a:hover {
   grid-template-columns: 1.1fr 0.9fr;
   gap: 20px;
   padding: 26px;
+  min-width: 0;
 }
 
 .hero__frame::after {
@@ -197,6 +198,7 @@ a:hover {
 .hero__copy {
   position: relative;
   z-index: 1;
+  min-width: 0;
 }
 
 .kicker {
@@ -313,6 +315,7 @@ a:hover {
   z-index: 1;
   display: grid;
   gap: 14px;
+  min-width: 0;
 }
 
 .shot {
@@ -356,6 +359,8 @@ a:hover {
   background: rgba(255, 255, 255, 0.03);
   padding: 20px 18px;
   box-shadow: var(--shadow2);
+  overflow: hidden;
+  min-width: 0;
 }
 
 .productCard__head {
@@ -411,6 +416,7 @@ a:hover {
   border-radius: 16px;
   border: 1px solid rgba(255, 255, 255, 0.1);
   background: rgba(0, 0, 0, 0.35);
+  overflow: hidden;
 }
 
 pre {
@@ -736,13 +742,139 @@ code {
 }
 
 @media (max-width: 640px) {
+  .shell {
+    padding: 20px 12px 60px;
+  }
+  .top {
+    padding: 8px 2px 12px;
+  }
+  .nav a {
+    padding: 6px 8px;
+    font-size: 11px;
+  }
+  .hero {
+    padding: 6px 0 0;
+  }
+  .hero__frame {
+    padding: 18px 14px;
+    border-radius: 20px;
+    gap: 16px;
+  }
+  .title {
+    font-size: clamp(26px, 7vw, 36px);
+  }
+  .lede {
+    font-size: 15px;
+  }
+  .quickRow {
+    padding: 10px;
+  }
+  .quickRow__code code {
+    font-size: 12px;
+    overflow-x: auto;
+    white-space: nowrap;
+    -webkit-overflow-scrolling: touch;
+  }
+  .quickRow__actions {
+    flex-direction: column;
+  }
+  .quickRow__actions .btn {
+    width: 100%;
+  }
+  .ctaRow {
+    flex-direction: column;
+  }
+  .ctaRow .btn {
+    width: 100%;
+    text-align: center;
+  }
+  .productGrid {
+    gap: 14px;
+    padding: 0 2px;
+  }
+  .productCard {
+    padding: 16px 14px;
+    border-radius: 20px;
+  }
+  .productCard h2 {
+    font-size: 19px;
+  }
+  pre {
+    font-size: 11.5px;
+    padding: 12px 10px;
+  }
   .capGrid {
     grid-template-columns: 1fr;
+    gap: 12px;
+    padding: 0 2px;
   }
-  .more__grid {
-    grid-template-columns: 1fr;
+  .capCard {
+    padding: 14px;
+    border-radius: 14px;
+  }
+  .flow {
+    padding: 0 2px;
+  }
+  .flow__card {
+    padding: 16px 14px;
+    border-radius: 20px;
   }
   .flowStep {
     grid-template-columns: 1fr;
+    padding: 10px;
+  }
+  .more {
+    padding: 0 2px;
+  }
+  .more__card {
+    padding: 16px 14px;
+    border-radius: 20px;
+  }
+  .more__grid {
+    grid-template-columns: 1fr;
+    gap: 8px;
+  }
+  .callout {
+    font-size: 14px;
+  }
+  .shot {
+    border-radius: 14px;
+  }
+}
+
+@media (max-width: 380px) {
+  .shell {
+    padding: 14px 8px 48px;
+  }
+  .brand__word {
+    font-size: 16px;
+  }
+  .nav {
+    gap: 4px;
+  }
+  .nav a {
+    padding: 5px 6px;
+    font-size: 10px;
+    letter-spacing: 0.04em;
+  }
+  .hero__frame {
+    padding: 14px 10px;
+  }
+  .title {
+    font-size: 24px;
+  }
+  .lede {
+    font-size: 14px;
+  }
+  .btn {
+    padding: 10px 12px;
+    font-size: 11px;
+  }
+  .productCard {
+    padding: 14px 12px;
+  }
+  .footer {
+    flex-direction: column;
+    gap: 6px;
   }
 }

package/docs/cache.md

@@ -41,13 +41,14 @@ Separate file cache for downloaded media (yt-dlp, direct media URLs). This is **
 - **Summaries**
   - key: `sha256({contentHash, promptHash, model, length, language, formatVersion})`
   - cache hit even if URL differs (content hash wins).
+  - `contentHash` comes from the `<content>` block actually sent to the model (so slide timelines / transcript extras affect the key).
 - **Slides** (manifest + on-disk images in the slides output dir)
   - key: `sha256({url, slideSettings, formatVersion})`
 
 ## Keys / hashes
 
 - `sha256` from Node `crypto` / Bun `crypto`.
-- `contentHash` from normalized extracted content.
+- `contentHash` from normalized prompt `<content>` (fallback: normalized extracted content when no tagged block exists).
 - `promptHash` from instruction block (custom prompt or default).
 - `formatVersion`: hardcoded constant to invalidate on prompt format changes.
 

package/docs/chrome-extension.md

@@ -32,7 +32,8 @@ Dev (repo checkout):
 - Use: `pnpm summarize daemon install --token <TOKEN> --dev` (autostart service runs `src/cli.ts` via `tsx`, no `dist/` build required).
 - E2E (Playwright): `pnpm -C apps/chrome-extension test:e2e`
   - First run: `pnpm -C apps/chrome-extension exec playwright install chromium`
-  - Headless: `HEADLESS=1 pnpm -C apps/chrome-extension test:e2e` (headful is more reliable for extensions)
+  - Chromium runs headless by default.
+  - Visible debugging: `SHOW_UI=1 pnpm -C apps/chrome-extension test:e2e` or `HEADLESS=0 pnpm -C apps/chrome-extension test:e2e`
 
 ## Troubleshooting
 
@@ -49,8 +50,10 @@ Dev (repo checkout):
 - “Stream ended unexpectedly” / empty chat response:
   - The daemon likely stopped mid-stream. Restart it, then click “Try again”.
   - `summarize daemon restart`
+- Slide strip/gallery missing after a parallel slide run failure:
+  - Click the slide notice “Try again” button. If the dedicated slide run never started, the extension now requests a fresh summarize+slides run instead of reusing the summary-only run.
 - Tweet video not transcribing / no progress:
-  - Ensure `yt-dlp` is available on your PATH (or set `YT_DLP_PATH`) and you have a transcription provider (`whisper.cpp` installed or `OPENAI_API_KEY` / `FAL_KEY`).
+  - Ensure `yt-dlp` is available on your PATH (or set `YT_DLP_PATH`) and you have a transcription provider (`whisper.cpp` installed or `GROQ_API_KEY` / `ASSEMBLYAI_API_KEY` / `GEMINI_API_KEY` / `OPENAI_API_KEY` / `FAL_KEY`).
   - Re-run `summarize daemon install --token <TOKEN>` to refresh the daemon env snapshot (launchd won’t inherit your shell PATH).
 - “Could not establish connection / Receiving end does not exist”:
   - The content script wasn’t injected (yet), or Chrome blocked site access.
@@ -105,6 +108,11 @@ See `docs/media.md` for detection and transcript rules.
 
 - The slides toggle lights up on media-friendly URLs (YouTube/watch|shorts, youtu.be, direct media) or when the page reports video/audio. Defaults to Video on those pages.
 - Turning slides **on** refreshes the current summary and requests slide extraction (`yt-dlp`, `ffmpeg`). OCR text is opt-in (Advanced setting) and requires `tesseract`. Missing tools surface a footer notice with install instructions; restart the daemon after installing.
+- Active slide mode is slide-first:
+  - vertical image/text cards
+  - transcript-first text; OCR fallback
+  - text can appear before slide images finish extracting
+  - the large summary block is hidden while slide cards are active
 - Slides stay off elsewhere and the toggle is disabled on non-media pages.
 
 ## SPA Navigation
@@ -161,7 +169,7 @@ Problem: daemon must be secured; extension must discover and pair with it.
     - `summarize daemon install --token <TOKEN>` (macOS: LaunchAgent, Linux: systemd user, Windows: Scheduled Task)
     - `summarize daemon status`
   - “Copy command” button.
-- Daemon stores token in `~/.summarize/daemon.json`.
+- Daemon stores paired tokens in `~/.summarize/daemon.json`.
 - Extension stores token in `chrome.storage.local`.
 - If daemon unreachable or 401: show Setup state + troubleshooting.
 
@@ -245,7 +253,7 @@ Notes:
   - `summarize daemon run` (foreground; used by autostart service)
 - Ensure “single daemon”:
   - Stable service name + predictable unit/task path
-  - `install` replaces previous install and validates token match
+  - `install` reuses the same daemon service and appends new tokens instead of invalidating older paired browsers
 
 Platform details:
 

package/docs/cli.md

@@ -12,7 +12,7 @@ Summarize can use installed CLIs (Claude, Codex, Gemini, Cursor Agent) as local
 
 - `cli/claude/<model>` (e.g. `cli/claude/sonnet`)
 - `cli/codex/<model>` (e.g. `cli/codex/gpt-5.2`)
-- `cli/gemini/<model>` (e.g. `cli/gemini/gemini-3-flash-preview`)
+- `cli/gemini/<model>` (e.g. `cli/gemini/gemini-3-flash`)
 - `cli/agent/<model>` (e.g. `cli/agent/gpt-5.2`)
 
 Use `--cli [provider]` (case-insensitive) for the provider default, or `--model cli/<provider>/<model>` to pin a model.
@@ -95,7 +95,7 @@ path-based prompt and enables the required tool flags:
       "order": ["claude", "gemini", "codex", "agent"]
     },
     "codex": { "model": "gpt-5.2" },
-    "gemini": { "model": "gemini-3-flash-preview", "extraArgs": ["--verbose"] },
+    "gemini": { "model": "gemini-3-flash", "extraArgs": ["--verbose"] },
     "claude": {
       "model": "sonnet",
       "binary": "/usr/local/bin/claude",

package/docs/config.md

@@ -56,7 +56,7 @@ For UI theme:
 
 ```json
 {
-  "model": { "id": "google/gemini-3-flash-preview" },
+  "model": { "id": "google/gemini-3-flash" },
   "env": { "OPENAI_API_KEY": "sk-..." },
   "output": { "language": "auto" },
   "prompt": "Explain like I am five.",
@@ -68,7 +68,7 @@ Shorthand (equivalent):
 
 ```json
 {
-  "model": "google/gemini-3-flash-preview"
+  "model": "google/gemini-3-flash"
 }
 ```
 
@@ -107,6 +107,7 @@ Set any env var in config (process env still wins):
 ```json
 {
   "env": {
+    "ASSEMBLYAI_API_KEY": "...",
     "OPENAI_API_KEY": "sk-...",
     "OPENROUTER_API_KEY": "sk-or-...",
     "FIRECRAWL_API_KEY": "...",
@@ -121,6 +122,8 @@ Legacy shortcut (still supported):
 {
   "apiKeys": {
     "openai": "sk-...",
+    "groq": "gsk-...",
+    "assemblyai": "...",
     "anthropic": "sk-ant-...",
     "google": "...",
     "openrouter": "sk-or-...",
@@ -260,7 +263,7 @@ For auto selection with rules:
     "rules": [
       {
         "when": ["video"],
-        "candidates": ["google/gemini-3-flash-preview"]
+        "candidates": ["google/gemini-3-flash"]
       },
       {
         "when": ["website", "youtube"],
@@ -359,9 +362,11 @@ Override API endpoints for any provider to use proxies, gateways, or compatible
 ```json
 {
   "openai": { "baseUrl": "https://my-openai-proxy.example.com/v1" },
+  "nvidia": { "baseUrl": "https://integrate.api.nvidia.com/v1" },
   "anthropic": { "baseUrl": "https://my-anthropic-proxy.example.com" },
   "google": { "baseUrl": "https://my-google-proxy.example.com" },
-  "xai": { "baseUrl": "https://my-xai-proxy.example.com" }
+  "xai": { "baseUrl": "https://my-xai-proxy.example.com" },
+  "zai": { "baseUrl": "https://api.zhipuai.cn/paas/v4" }
 }
 ```
 
@@ -370,6 +375,8 @@ Or via environment variables (which take precedence over config):
 | Provider  | Environment Variable(s)                      |
 | --------- | -------------------------------------------- |
 | OpenAI    | `OPENAI_BASE_URL`                            |
+| NVIDIA    | `NVIDIA_BASE_URL`                            |
 | Anthropic | `ANTHROPIC_BASE_URL`                         |
 | Google    | `GOOGLE_BASE_URL` (alias: `GEMINI_BASE_URL`) |
 | xAI       | `XAI_BASE_URL`                               |
+| Z.AI      | `Z_AI_BASE_URL` (alias: `ZAI_BASE_URL`)      |