@steipete/summarize 0.8.2 → 0.10.0

package/docs/assets/site.js

@@ -0,0 +1,72 @@
+const canonicalHost = 'summarize.sh'
+const redirectHostnames = new Set(['summarize.is', 'www.summarize.is'])
+
+const maybeRedirect = () => {
+  try {
+    const { hostname, pathname, search, hash } = window.location
+    if (!redirectHostnames.has(hostname)) return
+    const target = `https://${canonicalHost}${pathname}${search}${hash}`
+    window.location.replace(target)
+  } catch {
+    // ignore
+  }
+}
+
+const highlightNav = () => {
+  const path = window.location.pathname
+  const isDocs = path.includes('/docs/')
+  const navDocs = document.querySelector('a[data-nav="docs"]')
+  const navHome = document.querySelector('a[data-nav="home"]')
+  if (navDocs && isDocs) navDocs.setAttribute('aria-current', 'page')
+  if (navHome && !isDocs) navHome.setAttribute('aria-current', 'page')
+
+  const sideLinks = document.querySelectorAll('.side a[href]')
+  for (const a of sideLinks) {
+    const href = a.getAttribute('href') ?? ''
+    if (!href) continue
+    const normalized = href.replace(/^\.\//, '')
+    if (path.endsWith(normalized)) a.setAttribute('aria-current', 'page')
+  }
+}
+
+const wireCopyButtons = () => {
+  const buttons = document.querySelectorAll('[data-copy]')
+  const handleCopyClick = async (button) => {
+    const selector = button.getAttribute('data-copy')
+    const target = selector ? document.querySelector(selector) : null
+    const text = target?.textContent?.trim() ?? ''
+    if (!text) return
+    try {
+      await navigator.clipboard.writeText(text)
+      const prev = button.textContent ?? ''
+      button.textContent = 'Copied'
+      button.setAttribute('data-copied', '1')
+      window.setTimeout(() => {
+        button.textContent = prev
+        button.removeAttribute('data-copied')
+      }, 900)
+    } catch {
+      // ignore
+    }
+  }
+  for (const button of buttons) {
+    button.addEventListener('click', () => {
+      void handleCopyClick(button)
+    })
+  }
+}
+
+const reveal = () => {
+  const items = document.querySelectorAll('.reveal')
+  let i = 0
+  for (const el of items) {
+    const delay = Math.min(380, i * 70)
+    window.setTimeout(() => el.classList.add('is-on'), delay)
+    i++
+  }
+}
+
+maybeRedirect()
+highlightNav()
+wireCopyButtons()
+reveal()

package/docs/cache.md

@@ -21,15 +21,28 @@ Lightweight, CLI-only SQLite cache. Single DB file.
 - Override: `cache.path` in config.
 - SQLite pragmas: WAL, NORMAL sync, busy timeout, incremental vacuum.
 
+## Media cache (downloads)
+
+Separate file cache for downloaded media (yt-dlp, direct media URLs). This is **not** the SQLite DB.
+
+- Default path: `~/.summarize/cache/media`
+- TTL: 7 days
+- Size cap: 2048 MB
+- Config: `cache.media` (see below)
+- CLI: `--no-media-cache` disables media caching only
+- Note: `--no-cache` **does not** disable the media cache
+
 ## What we cache
 
 - **Transcripts**
-  - key: `sha256({url, namespace, formatVersion})`
+  - key: `sha256({url, namespace, fileMtime?, formatVersion})` (local file paths include `fileMtime` for invalidation)
 - **Extracted content** (URL → text/markdown)
   - key: `sha256({url, extractSettings, formatVersion})`
 - **Summaries**
   - key: `sha256({contentHash, promptHash, model, length, language, formatVersion})`
   - cache hit even if URL differs (content hash wins).
+- **Slides** (manifest + on-disk images in the slides output dir)
+  - key: `sha256({url, slideSettings, formatVersion})`
 
 ## Keys / hashes
 
@@ -46,7 +59,14 @@ Lightweight, CLI-only SQLite cache. Single DB file.
     "enabled": true,
     "maxMb": 512,
     "ttlDays": 30,
-    "path": "~/.summarize/cache.sqlite"
+    "path": "~/.summarize/cache.sqlite",
+    "media": {
+      "enabled": true,
+      "maxMb": 2048,
+      "ttlDays": 7,
+      "path": "~/.summarize/cache/media",
+      "verify": "size"
+    }
   }
 }
 ```
@@ -55,7 +75,7 @@ Defaults: `enabled=true`, `maxMb=512`, `ttlDays=30`, `path` unset.
 
 ## CLI flags
 
-- `--no-cache`: bypass read + write.
+- `--no-cache`: bypass summary cache reads + writes (LLM output). Extract/transcript caches still apply.
 - `--cache-stats`: print cache stats and exit.
 - `--clear-cache`: delete cache DB (and WAL/SHM); must be used alone.
 
@@ -65,6 +85,12 @@ Defaults: `enabled=true`, `maxMb=512`, `ttlDays=30`, `path` unset.
 - Size cap: if DB > `maxMb`, delete oldest entries by `last_accessed_at` until under cap.
 - Optional count cap if needed later.
 
+Media cache eviction:
+
+- TTL sweep on read/write.
+- Size cap: evict least-recently-used files until under cap.
+- `verify` controls integrity checks: `size` (default), `hash`, or `none`.
+
 ## Notes
 
 - No extension cache (daemon uses CLI cache).

package/docs/chrome-extension.md

@@ -4,7 +4,7 @@ read_when:
   - "When working on the extension, daemon, or side panel UX."
 ---
 
-# Chrome Side Panel (Chrome Extension + Daemon)
+# Browser Side Panel (Chrome + Firefox Extension + Daemon)
 
 Goal: Chrome **Side Panel** (“real sidebar”) summarizes **what you see** on the current tab. Panel open → navigation → auto summarize (optional) → **streaming** Markdown rendered in-panel.
 
@@ -14,12 +14,18 @@ Quickstart:
   - `npm i -g @steipete/summarize`
   - `brew install steipete/tap/summarize` (macOS arm64)
 - Build/load extension: `apps/chrome-extension/README.md`
+- Firefox sidebar build: `pnpm -C apps/chrome-extension build:firefox` (load via `about:debugging` → temporary add-on)
 - Open side panel → copy token install command → run:
   - `summarize daemon install --token <TOKEN>` (macOS: LaunchAgent, Linux: systemd user, Windows: Scheduled Task)
 - Verify:
   - `summarize daemon status`
   - Restart (if needed): `summarize daemon restart`
 
+Firefox notes:
+- Sidebar UX differs from Chrome’s side panel (persistent sidebar instead of slide-in panel).
+- Firefox testing is limited in Playwright; see `apps/chrome-extension/tests/README-firefox.md`.
+- Compatibility details: `apps/chrome-extension/docs/firefox.md`.
+
 Dev (repo checkout):
 
 - Use: `pnpm summarize daemon install --token <TOKEN> --dev` (autostart service runs `src/cli.ts` via `tsx`, no `dist/` build required).
@@ -32,12 +38,23 @@ Dev (repo checkout):
 - “Daemon not reachable”:
   - `summarize daemon status`
   - Logs: `~/.summarize/logs/daemon.err.log`
+- “Need extension-side traces”:
+  - Options → Logs → `extension.log` (panel/background events).
+  - Enable “Extended logging” in Advanced settings for full pipeline traces.
+- “Stream ended unexpectedly” / empty chat response:
+  - The daemon likely stopped mid-stream. Restart it, then click “Try again”.
+  - `summarize daemon restart`
 - 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`).
   - 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.
   - Chrome → extension details → “Site access” → “On all sites” (or allow the domain), then reload the tab.
+- “<site> wants to look for and connect to any device on your local network”:
+  - Trigger: content scripts (page context) hitting the daemon on `http://127.0.0.1:8787` (hover summaries) can cause Chrome to attribute the request to the current origin and prompt per-site.
+  - Fix: hover summaries must proxy daemon calls via the extension background service worker (reload the extension after updating).
+  - Verify daemon: `summarize daemon status` (or `curl http://127.0.0.1:8787/health`).
+  - Repro/dev: `pnpm -C apps/chrome-extension dev` then enable “Hover summaries” and hover a link.
 
 ## Architecture
 
@@ -73,6 +90,18 @@ The daemon decides the best pipeline:
 - Normal articles with extracted text → prefer **page** pipeline (“what you see”).
 - Fallback: if the preferred path fails before output starts, try the other input (when available).
 
+## Video Selection (Page vs Video)
+
+When the page contains embedded audio/video, the Summarize button gains a dropdown caret. Click the caret to pick Page vs Video/Audio. Selecting Video/Audio forces URL mode with transcript-first extraction (captions → yt-dlp/Whisper fallback). Selection is per-run (not persisted).
+
+See `docs/media.md` for detection and transcript rules.
+
+## Slides (Side Panel)
+
+- 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.
+- Slides stay off elsewhere and the toggle is disabled on non-media pages.
+
 ## SPA Navigation
 
 - Background listens to `chrome.webNavigation.onHistoryStateUpdated` (SPA route changes) and `tabs.onUpdated` (page loads).
@@ -85,23 +114,36 @@ The daemon decides the best pipeline:
 - `linkify: true`.
 - Render links with `target=_blank` + `rel=noopener noreferrer`.
 
+## Timestamp Links (Chat)
+
+- When timed transcripts are available, chat context includes `[mm:ss]` lines.
+- Assistant is prompted to cite timestamps; clicking them seeks the current media (video/audio) while preserving play/pause state.
+
 ## Model Selection UX
 
 - Settings:
   - Model preset (Options → Advanced): `auto` | `free` | custom string (e.g. `openai/gpt-5-mini`, `openrouter/...`).
-  - Length: `short|medium|long|xl|xxl` (or a character target like `20k`).
+  - Length: `short|medium|long|xl|xxl` (or a character target like `20k`). Tooltips show target ranges + paragraph guidance (from `packages/core/src/prompts/summary-lengths.ts`).
   - Language: `auto` (match source) or a tag like `en`, `de`, `pt-BR` (or free-form like “German”).
   - Prompt override (advanced): custom instruction prefix (context + content still appended).
   - Auto summarize: on/off.
+  - Hover summaries: on/off (side panel drawer, default off).
   - Typography: font family (dropdown + custom), font size (slider).
-- Advanced overrides (collapsed by default; click the section title to expand).
+- Advanced overrides (Options → Advanced tab).
   - Leave blank to use daemon config/defaults; set a value to override.
+  - Chat (advanced): enable/disable the side panel chat input (default on; summary is the first message).
+  - Summary timestamps (advanced): include `[mm:ss]` links in summaries for media when available (default on).
+  - Slides parallel (advanced): show summary first and extract slides in parallel (default on).
+  - Slides OCR text (advanced): allow OCR text as a slide text source (default off).
+  - Extended logging: send full input/output to daemon logs (requires daemon logging enabled).
+  - Hover summary prompt: customize the prompt used for link hover summaries (prefilled; reset to default).
   - Pipeline mode: `page|url` (default auto).
   - Firecrawl: `off|auto|always`.
   - Markdown mode: `readability|llm|auto|off`.
-  - Preprocess: `off|auto|always`.
-  - YouTube mode: `no-auto|yt-dlp|web|apify` (default auto).
-  - Timeout (e.g. `90s`, `2m`), retries, max output tokens (e.g. `2k`).
+- Preprocess: `off|auto|always`.
+- YouTube mode: `no-auto|yt-dlp|web|apify` (default auto).
+- Timeout (e.g. `90s`, `2m`), retries, max output tokens (e.g. `2k`).
+- Process manager: live list of daemon-spawned tools (ffmpeg, yt-dlp, tesseract, etc.) with logs.
 - Extension includes current settings in request; daemon treats them like CLI flags (`--model`, `--length`, `--language`, `--prompt`).
 
 ## Token Pairing / Setup Mode
@@ -134,10 +176,46 @@ Problem: daemon must be secured; extension must discover and pair with it.
     - `language?: string` (e.g. `auto`, `en`, `de`, `pt-BR`)
     - `prompt?: string` (custom instruction prefix)
     - `mode?: "auto" | "page" | "url"` (default: `"auto"`)
-    - `maxCharacters?: number | null` (caps URL-mode extraction before summarization)
+    - `maxCharacters?: number | null` (caps URL-mode extraction before summarization; ignored for extract-only unless explicitly provided)
+    - `format?: "text" | "markdown"` (default: `"text"`)
+    - `markdownMode?: "readability" | "auto" | "llm" | "off"` (only when `format: "markdown"`)
+    - `preprocess?: "off" | "auto" | "always"` (markitdown/HTML preprocess)
+    - `extractOnly?: boolean` (when `true`, returns extracted content without summarizing; requires `mode: "url"`)
     - `text?: string` (required for `mode: "page"`; optional for `auto`)
     - `truncated?: boolean` (optional; indicates extracted `text` was shortened)
   - 200 JSON: `{ ok: true, id }`
+- `GET /v1/summarize/<id>/slides/events`
+  - Headers: `Authorization: Bearer <token>`
+  - SSE stream of slide updates (`slides`, `status`, `done`, `error`) independent of summary stream.
+- `POST /v1/agent` (SSE by default; JSON via `Accept: application/json` or `?format=json`)
+  - Headers: `Authorization: Bearer <token>`
+  - Body:
+    - `url: string` (required)
+    - `title?: string | null`
+    - `pageContent: string`
+    - `cacheContent?: string` (used for cache key; defaults to `pageContent`)
+    - `messages: Array<Message>` (pi-ai format)
+    - `model?: string`
+    - `length?: string` (e.g. `short`, `xl`, `20k`)
+    - `language?: string` (e.g. `auto`, `en`, `de`)
+    - `tools?: string[]`
+    - `automationEnabled?: boolean`
+  - SSE events:
+    - `event: chunk` `data: { text }`
+    - `event: assistant` `data: { ...assistant }`
+    - `event: done` `data: {}`
+    - `event: error` `data: { message }`
+- `POST /v1/agent/history`
+  - Headers: `Authorization: Bearer <token>`
+  - Body:
+    - `url: string` (required)
+    - `pageContent: string`
+    - `cacheContent?: string` (used for cache key; defaults to `pageContent`)
+    - `model?: string`
+    - `length?: string`
+    - `language?: string`
+    - `automationEnabled?: boolean`
+  - 200 JSON: `{ ok: true, messages }`
 - `GET /v1/summarize/:id/events` (SSE)
   - `event: chunk` `data: { text }`
   - `event: meta` `data: { model }`

package/docs/config.md

@@ -36,6 +36,13 @@ For prompt:
 1. CLI flag `--prompt` / `--prompt-file`
 2. Config file `prompt`
 3. Built-in default prompt
+
+For UI theme:
+
+1. CLI flag `--theme`
+2. Env `SUMMARIZE_THEME`
+3. Config file `ui.theme`
+4. Built-in default (`aurora`)
 ## Format
 
 `~/.summarize/config.json`:
@@ -44,7 +51,8 @@ For prompt:
 {
   "model": { "id": "google/gemini-3-flash-preview" },
   "output": { "language": "auto" },
-  "prompt": "Explain like I am five."
+  "prompt": "Explain like I am five.",
+  "ui": { "theme": "ember" }
 }
 ```
 
@@ -94,11 +102,75 @@ Configure the on-disk SQLite cache (extracted content, transcripts, summaries).
     "enabled": true,
     "maxMb": 512,
     "ttlDays": 30,
-    "path": "~/.summarize/cache.sqlite"
+    "path": "~/.summarize/cache.sqlite",
+    "media": {
+      "enabled": true,
+      "maxMb": 2048,
+      "ttlDays": 7,
+      "path": "~/.summarize/cache/media",
+      "verify": "size"
+    }
+  }
+}
+```
+
+Notes:
+
+- `cache.media` controls the **media file** cache (yt-dlp downloads).
+- `--no-cache` bypasses summary caching only (LLM output); extract/transcript caches still apply. Use `--no-media-cache` for media.
+- `verify`: `size` (default), `hash`, or `none`.
+
+## UI theme
+
+Set a default CLI theme:
+
+```json
+{
+  "ui": { "theme": "moss" }
+}
+```
+
+## Slides defaults
+
+Enable slides by default and tune extraction parameters:
+
+```json
+{
+  "slides": {
+    "enabled": true,
+    "ocr": false,
+    "dir": "slides",
+    "sceneThreshold": 0.3,
+    "max": 20,
+    "minDuration": 2
   }
 }
 ```
 
+## Logging (daemon)
+
+Enable JSON log files for the daemon:
+
+```json
+{
+  "logging": {
+    "enabled": true,
+    "level": "info",
+    "format": "json",
+    "file": "~/.summarize/logs/daemon.jsonl",
+    "maxMb": 10,
+    "maxFiles": 3
+  }
+}
+```
+
+Notes:
+
+- Default: logging is off.
+- `format`: `json` (default) or `pretty`.
+- `maxMb` is per file; `maxFiles` controls rotation (ring).
+- Extension “Extended logging” sends full input/output to daemon logs (large). Cache hits skip content logging.
+
 ## Presets
 
 Define presets you can select via `--model <preset>`:

package/docs/extract-only.md

@@ -13,13 +13,21 @@ Deprecated alias: `--extract-only`.
 ## Notes
 
 - No summarization LLM call happens in this mode.
-- `--format md` may still convert HTML → Markdown (depending on `--markdown-mode` and available tools).
+- No extraction cap is applied. Use `--max-extract-characters <count>` to cap output if needed.
+- `--format md` may still convert HTML to Markdown (depending on `--markdown-mode` and available tools).
 - `--length` is intended for summarization guidance; extraction prints full content.
+- `--timestamps` keeps the plain transcript text but also exposes `transcriptSegments` and `transcriptTimedText` (JSON) and prints a timed transcript block when available.
+- `--slides` runs slide detection (YouTube/direct video URLs). Slide metadata is included in JSON output and written to `slides.json` in the slide directory.
+  - When combined with `--extract` for videos that have timed transcripts, the CLI interleaves slide images inline at matching timestamps.
+  - Scene detection auto-tunes using sampled frame hashes.
 - For non-YouTube URLs with `--format md`, the CLI uses Readability article HTML as the default Markdown input (`--markdown-mode readability`).
   - Use `--markdown-mode auto` to prefer LLM/markitdown conversion without Readability preprocessing.
   - Use `--markdown-mode llm` to force an LLM conversion.
   - Use `--firecrawl always` to try Firecrawl first.
-- For non-YouTube URLs with `--format md`, `--markdown-mode auto` can convert HTML → Markdown via an LLM when configured.
+- For non-YouTube URLs with `--format md`, `--markdown-mode auto` can convert HTML to Markdown via an LLM when configured.
   - Force it with `--markdown-mode llm`.
   - If no LLM is configured, `--markdown-mode auto` may fall back to `uvx markitdown` when available.
 - `--markdown-mode readability` uses Readability to extract article HTML before Markdown conversion.
+
+Daemon note:
+- `/v1/summarize` supports `format: "markdown"` + `markdownMode` for extract-only output (use `extractOnly: true`).

package/docs/index.html

@@ -0,0 +1,205 @@
+<!doctype html>
+<html lang="en">
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width,initial-scale=1" />
+    <meta name="color-scheme" content="dark light" />
+    <title>summarize — CLI + Chrome Side Panel for fast summaries</title>
+    <meta
+      name="description"
+      content="Summarize is a fast CLI and a Chrome Side Panel extension for clean extraction and sharp summaries."
+    />
+    <link rel="canonical" href="https://summarize.sh/" />
+    <meta property="og:title" content="summarize" />
+    <meta property="og:description" content="Fast summaries, in the CLI and the Chrome Side Panel." />
+    <meta property="og:type" content="website" />
+    <meta property="og:url" content="https://summarize.sh/" />
+    <meta name="theme-color" content="#0b0f12" />
+
+    <link rel="preconnect" href="https://fonts.googleapis.com" />
+    <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin />
+    <link
+      href="https://fonts.googleapis.com/css2?family=Fraunces:wght@600;700&family=IBM+Plex+Mono:wght@400;500;600&family=IBM+Plex+Sans:wght@400;500;600&display=swap"
+      rel="stylesheet"
+    />
+    <link rel="stylesheet" href="./assets/site.css" />
+  </head>
+  <body>
+    <main class="shell">
+      <header class="top">
+        <a class="brand" href="./index.html" aria-label="summarize home">
+          <span class="brand__mark" aria-hidden="true">s</span>
+          <span class="brand__word">summarize</span>
+        </a>
+        <nav class="nav" aria-label="Primary">
+          <a data-nav="home" href="./index.html">Home</a>
+          <a data-nav="docs" href="./docs/index.html">Docs</a>
+          <a href="https://github.com/steipete/summarize">GitHub</a>
+        </nav>
+      </header>
+
+      <section class="hero">
+        <div class="hero__frame reveal">
+          <div class="hero__copy">
+            <p class="kicker">CLI + Chrome Extension</p>
+            <h1 class="title">Summaries that live where you work.</h1>
+            <p class="lede">
+              Summarize turns links, files, and media into sharp summaries with a real extraction pipeline.
+              Use the <strong>CLI</strong> for automation or the <strong>Chrome Side Panel</strong> for one-click
+              summaries of the current tab. Supports <strong>local, paid, and free</strong> models.
+            </p>
+            <div class="ctaRow">
+              <a class="btn btn--primary" href="./docs/index.html">Docs</a>
+              <a class="btn" href="https://github.com/steipete/summarize">GitHub</a>
+            </div>
+            <div class="quickRow">
+              <div class="quickRow__label">Quickstart</div>
+              <div class="quickRow__code">
+                <code id="install">npm i -g @steipete/summarize</code>
+                <code id="run">summarize "https://example.com/article"</code>
+              </div>
+              <div class="quickRow__actions">
+                <button class="btn" data-copy="#install">Copy install</button>
+                <button class="btn" data-copy="#run">Copy run</button>
+              </div>
+            </div>
+          </div>
+
+          <div class="hero__visual">
+            <figure class="shot shot--cli">
+              <img src="./assets/summarize-cli.png" alt="Summarize CLI output" loading="lazy" />
+              <figcaption>CLI: streaming summaries in the terminal.</figcaption>
+            </figure>
+            <figure class="shot shot--ext">
+              <img src="./assets/summarize-extension.png" alt="Summarize Chrome extension side panel" loading="lazy" />
+              <figcaption>Extension: a live Side Panel for the active tab.</figcaption>
+            </figure>
+          </div>
+        </div>
+      </section>
+
+      <section class="productGrid" aria-label="Products">
+        <article class="productCard reveal" id="cli">
+          <div class="productCard__head">
+            <span class="badge">CLI</span>
+            <h2>Fast summaries, scripted or interactive.</h2>
+          </div>
+          <p>
+            Built for automation: extract clean text, summarize with your model, and output JSON or Markdown.
+            Works with URLs, PDFs, images, audio/video, YouTube, and podcasts.
+          </p>
+          <ul class="list">
+            <li><strong>Extract + summarize</strong> with Firecrawl fallback.</li>
+            <li><strong>Media pipeline</strong> with transcript-first flow and Whisper fallback.</li>
+            <li><strong>Scriptable output</strong> via <code>--json</code>, <code>--extract</code>, <code>--metrics</code>.</li>
+          </ul>
+          <div class="cardCode">
+            <pre><code>summarize "https://example.com" --length long
+summarize "https://youtu.be/..." --youtube auto
+summarize "/path/report.pdf" --model google/gemini-3-flash-preview</code></pre>
+          </div>
+          <div class="ctaRow">
+            <a class="btn btn--ghost" href="./docs/cli.html">CLI docs</a>
+            <a class="btn" href="./docs/config.html">Config</a>
+          </div>
+        </article>
+
+        <article class="productCard reveal" id="extension">
+          <div class="productCard__head">
+            <span class="badge badge--alt">Chrome Extension</span>
+            <h2>Summaries in the Side Panel, one click away.</h2>
+          </div>
+          <p>
+            A real Chrome Side Panel with a tiny local daemon. It streams Markdown summaries for the active tab,
+            with auto-summary on navigation.
+          </p>
+          <ol class="steps">
+            <li>Install the CLI + daemon.</li>
+            <li>Load the unpacked extension.</li>
+            <li>Open Side Panel and connect with the token.</li>
+          </ol>
+          <div class="callout">
+            <strong>Runs locally</strong> on your machine. The daemon is localhost-only and token-protected.
+          </div>
+          <div class="ctaRow">
+            <a class="btn btn--ghost" href="./docs/chrome-extension.html">Extension docs</a>
+            <a class="btn" href="https://github.com/steipete/summarize/releases/latest">Latest release</a>
+          </div>
+        </article>
+      </section>
+
+      <section class="capGrid" aria-label="Capabilities">
+        <div class="capCard reveal">
+          <span class="capCard__eyebrow">Extraction</span>
+          <h3>HTML -> clean text -> summary</h3>
+          <p>Readability, markitdown, and Firecrawl fallback when sites fight back.</p>
+        </div>
+        <div class="capCard reveal">
+          <span class="capCard__eyebrow">Media</span>
+          <h3>Podcast + YouTube aware</h3>
+          <p>Prefers published transcripts, then Whisper when needed.</p>
+        </div>
+        <div class="capCard reveal">
+          <span class="capCard__eyebrow">Models</span>
+          <h3>Provider-agnostic</h3>
+          <p>Local OpenAI-compatible gateways, paid providers, and OpenRouter free models.</p>
+        </div>
+        <div class="capCard reveal">
+          <span class="capCard__eyebrow">Outputs</span>
+          <h3>Readable + scriptable</h3>
+          <p>Streaming TTY output, ANSI Markdown, JSON diagnostics, and metrics.</p>
+        </div>
+      </section>
+
+      <section class="flow">
+        <div class="flow__card reveal">
+          <h2>How it works</h2>
+          <div class="flow__steps">
+            <div class="flowStep">
+              <span class="flowStep__num">01</span>
+              <div>
+                <h4>Fetch + extract</h4>
+                <p>Pull the source, clean HTML, normalize, or convert to Markdown.</p>
+              </div>
+            </div>
+            <div class="flowStep">
+              <span class="flowStep__num">02</span>
+              <div>
+                <h4>Transcript when needed</h4>
+                <p>Use published transcripts, then Whisper fallback for media.</p>
+              </div>
+            </div>
+            <div class="flowStep">
+              <span class="flowStep__num">03</span>
+              <div>
+                <h4>Summarize + format</h4>
+                <p>LLM output, streaming in the CLI or Side Panel, with metrics.</p>
+              </div>
+            </div>
+          </div>
+        </div>
+      </section>
+
+      <section class="more">
+        <div class="more__card reveal">
+          <h2>More</h2>
+          <div class="more__grid">
+            <a class="moreLink" href="./docs/website.html">Website extraction</a>
+            <a class="moreLink" href="./docs/youtube.html">YouTube transcripts</a>
+            <a class="moreLink" href="./docs/media.html">Media pipeline</a>
+            <a class="moreLink" href="./docs/model-auto.html">Auto model rules</a>
+            <a class="moreLink" href="./docs/llm.html">Model + key setup</a>
+            <a class="moreLink" href="./docs/config.html">Config schema</a>
+          </div>
+        </div>
+      </section>
+
+      <footer class="footer">
+        <span>Canonical: <a href="https://summarize.sh">summarize.sh</a></span>
+        <span><a href="./docs/index.html">Docs</a> · <a href="https://github.com/steipete/summarize">GitHub</a></span>
+      </footer>
+    </main>
+
+    <script type="module" src="./assets/site.js"></script>
+  </body>
+</html>

package/docs/index.md

@@ -0,0 +1,25 @@
+---
+summary: "Docs index for summarize behaviors and modes."
+---
+
+# Docs
+
+- [Chrome extension](chrome-extension.md)
+- [Cache](cache.md)
+- [CLI](cli.md)
+- [Config](config.md)
+- [Extract](extract-only.md)
+- [Firecrawl](firecrawl.md)
+- [Language](language.md)
+- [LLM](llm.md)
+- [Manual tests](manual-tests.md)
+- [Model auto](model-auto.md)
+- [OpenAI](openai.md)
+- [Releasing](releasing.md)
+- [Smoke test](smoketest.md)
+- [Website](website.md)
+- [YouTube](youtube.md)
+
+## Website
+
+- Jekyll site source: `docs/` (Markdown → HTML via GitHub Pages Actions)