@steipete/summarize 0.7.0 → 0.8.0

package/docs/llm.md

@@ -1,3 +1,9 @@
+---
+summary: "LLM usage, env vars, flags, and prompt rules."
+read_when:
+  - "When changing model selection or prompt formatting."
+---
+
 # LLM / summarization mode
 
 By default `summarize` will call an LLM using **direct provider API keys**. When CLI tools are
@@ -12,6 +18,7 @@ installed, auto mode can use local CLI models when `cli.enabled` is set (see `do
 
 - `.env` (optional): when running the CLI, `summarize` also reads `.env` in the current working directory and merges it into the environment (real env vars win).
 - `XAI_API_KEY` (required for `xai/...` models)
+- `XAI_BASE_URL` (optional; override xAI API endpoint)
 - `OPENAI_API_KEY` (required for `openai/...` models)
 - `OPENAI_BASE_URL` (optional; OpenAI-compatible API endpoint, e.g. OpenRouter)
 - `OPENAI_USE_CHAT_COMPLETIONS` (optional; force OpenAI chat completions)
@@ -19,7 +26,9 @@ installed, auto mode can use local CLI models when `cli.enabled` is set (see `do
 - `Z_AI_API_KEY` (required for `zai/...` models; supports `ZAI_API_KEY` alias)
 - `Z_AI_BASE_URL` (optional; override default Z.AI base URL)
 - `GEMINI_API_KEY` (required for `google/...` models; also accepts `GOOGLE_GENERATIVE_AI_API_KEY` / `GOOGLE_API_KEY`)
+- `GOOGLE_BASE_URL` / `GEMINI_BASE_URL` (optional; override Google API endpoint)
 - `ANTHROPIC_API_KEY` (required for `anthropic/...` models)
+- `ANTHROPIC_BASE_URL` (optional; override Anthropic API endpoint)
 - `SUMMARIZE_MODEL` (optional; overrides default model selection)
 - `CLAUDE_PATH` / `CODEX_PATH` / `GEMINI_PATH` (optional; override CLI binary paths)
 
@@ -43,12 +52,23 @@ installed, auto mode can use local CLI models when `cli.enabled` is set (see `do
   - See `docs/model-auto.md`
 - `--model <preset>`
   - Uses a config-defined preset (see `docs/config.md` → “Presets”).
+- `--prompt <text>` / `--prompt-file <path>`
+  - Overrides the built-in summary instructions (prompt becomes the instruction prefix).
+  - Prompts are wrapped in `<instructions>`, `<context>`, `<content>` tags.
+  - When `--length` is numeric, we add `Output is X characters.` When `--language` is explicitly set, we add `Output should be <language>.`
+- `--no-cache`
+  - Bypass cache reads and writes (extracted content, transcripts, summaries).
+- `--cache-stats`
+  - Print cache stats and exit.
+- `--clear-cache`
+  - Delete the cache database and exit. Must be used alone.
 - `--video-mode auto|transcript|understand`
   - Only relevant for video inputs / video-only pages.
 - `--length short|medium|long|xl|xxl|<chars>`
   - This is *soft guidance* to the model (no hard truncation).
   - Minimum numeric value: 50 chars.
   - Default: `long`.
+  - Output format is Markdown; use short paragraphs and only add bullets when they improve scanability.
 - `--max-output-tokens <count>`
   - Hard cap for output tokens (optional).
   - If omitted, no max token parameter is sent (provider default).

package/docs/manual-tests.md

@@ -1,3 +1,9 @@
+---
+summary: "Manual end-to-end test checklist for model and input coverage."
+read_when:
+  - "When doing release validation."
+---
+
 # Manual tests
 
 Goal: sanity-check auto selection + presets end-to-end.

package/docs/model-auto.md

@@ -1,3 +1,9 @@
+---
+summary: "Auto model selection logic and config rules."
+read_when:
+  - "When changing auto selection behavior."
+---
+
 # Auto model selection (`--model auto`)
 
 `--model auto` picks a model based on input kind + token size, and retries with fallbacks when something fails.

package/docs/openai.md

@@ -1,3 +1,9 @@
+---
+summary: "OpenAI model usage and flags."
+read_when:
+  - "When changing OpenAI integration."
+---
+
 # OpenAI models
 
 Use OpenAI directly by choosing an `openai/...` model id.

package/docs/site/index.html

@@ -40,7 +40,8 @@
               <h1 class="title">Link → clean text → summary.</h1>
               <p class="lede">
                 A pragmatic pipeline for the web: <strong>extract</strong> the good stuff, <strong>sanitize</strong> it,
-                then <strong>summarize</strong> with your model of choice. Podcast episode pages included.
+                then <strong>summarize</strong> with your model of choice. Podcast episode pages included. Plus a Chrome
+                side panel extension when you want it in-browser.
               </p>
 
               <div class="pillRow">
@@ -48,6 +49,7 @@
                 <span class="pill"><span class="pill__dot" aria-hidden="true" style="background: var(--accent2)"></span> YouTube transcripts</span>
                 <span class="pill"><span class="pill__dot" aria-hidden="true" style="background: var(--accent3)"></span> Podcast compatibility</span>
                 <span class="pill"><span class="pill__dot" aria-hidden="true" style="background: var(--accent3)"></span> JSON + metrics</span>
+                <span class="pill"><span class="pill__dot" aria-hidden="true" style="background: var(--accent2)"></span> Chrome extension</span>
               </div>
 
               <div class="ctaRow">
@@ -118,6 +120,14 @@
           <p>OpenAI, xAI, Gemini — use what you’ve got, pin what you need.</p>
           <div class="small">Control via <code>--model</code> + env vars</div>
         </div>
+        <div class="card reveal">
+          <h2>Chrome side panel</h2>
+          <p>Summarize the active tab without leaving your browser.</p>
+          <div class="small">
+            Companion extension for the CLI. Download from the
+            <a href="https://github.com/steipete/summarize/releases/latest">latest GitHub release</a>.
+          </div>
+        </div>
       </section>
 
       <footer class="footer">

package/docs/smoketest.md

@@ -1,3 +1,9 @@
+---
+summary: "20-case smoke test plan for inputs and models."
+read_when:
+  - "When running smoke tests."
+---
+
 # Smoke Test Plan (20 combos)
 
 Goal: exercise URL + file inputs, extraction + LLM summary paths, multiple models.

package/docs/website.md

@@ -1,3 +1,9 @@
+---
+summary: "Website extraction pipeline, flags, and fallbacks."
+read_when:
+  - "When changing website extraction behavior."
+---
+
 # Website mode
 
 Use this for non-YouTube URLs.

package/docs/youtube.md

@@ -1,11 +1,18 @@
+---
+summary: "YouTube transcript extraction modes and fallbacks."
+read_when:
+  - "When changing YouTube handling."
+---
+
 # YouTube mode
 
 YouTube URLs use transcript-first extraction.
 
-## `--youtube auto|web|apify|yt-dlp`
+## `--youtube auto|web|no-auto|apify|yt-dlp`
 
-- `auto` (default): try `youtubei` → `captionTracks` → Apify (if token exists) → `yt-dlp` (if configured)
+- `auto` (default): try `youtubei` → `captionTracks` → `yt-dlp` (if configured) → Apify (if token exists)
 - `web`: try `youtubei` → `captionTracks` only
+- `no-auto`: try creator captions only (skip auto-generated/ASR) → `yt-dlp` (if configured)
 - `apify`: Apify only
 - `yt-dlp`: download audio + transcribe (local `whisper.cpp` preferred; OpenAI/FAL fallback)
 

package/package.json

@@ -1,11 +1,11 @@
 {
   "name": "@steipete/summarize",
-  "version": "0.7.0",
+  "version": "0.8.0",
   "description": "Link → clean text → summary.",
   "type": "module",
   "bin": {
-    "summarize": "./dist/cli.cjs",
-    "summarizer": "./dist/cli.cjs"
+    "summarize": "./dist/cli.js",
+    "summarizer": "./dist/cli.js"
   },
   "main": "./dist/esm/index.js",
   "module": "./dist/esm/index.js",
@@ -35,20 +35,16 @@
     "node": ">=22"
   },
   "dependencies": {
-    "@steipete/summarize-core": "0.7.0"
+    "@mariozechner/pi-ai": "^0.30.2",
+    "@steipete/summarize-core": "0.8.0"
   },
   "devDependencies": {
-    "@ai-sdk/anthropic": "3.0.1",
-    "@ai-sdk/google": "3.0.1",
-    "@ai-sdk/openai": "3.0.1",
-    "@ai-sdk/xai": "3.0.1",
     "@biomejs/biome": "^2.3.10",
     "@fal-ai/client": "^1.8.1",
     "@types/jsdom": "^27.0.0",
     "@types/node": "^25.0.3",
     "@types/sanitize-html": "^2.16.0",
     "@vitest/coverage-v8": "^4.0.16",
-    "ai": "6.0.3",
     "commander": "^14.0.2",
     "esbuild": "^0.27.2",
     "file-type": "^21.2.0",
@@ -72,6 +68,7 @@
     "build:cli": "node scripts/build-cli.mjs",
     "build:bun": "bun scripts/build-bun.js",
     "build:bun:test": "bun scripts/build-bun.js --test",
+    "docs:list": "tsx scripts/docs-list.ts",
     "typecheck": "tsc -p tsconfig.build.json --noEmit",
     "summarize": "tsx src/cli.ts",
     "s": "tsx src/cli.ts",