@jackwener/opencli 1.4.0 → 1.5.0

package/docs/adapters/browser/douyin.md

@@ -0,0 +1,75 @@
+# Douyin (抖音创作者中心)
+
+**Mode**: 🔐 Browser · **Domain**: `creator.douyin.com`
+
+## Commands
+
+| Command | Description |
+|---------|-------------|
+| `opencli douyin profile` | 获取账号信息 |
+| `opencli douyin videos` | 获取作品列表 |
+| `opencli douyin drafts` | 获取草稿列表 |
+| `opencli douyin draft` | 上传视频并保存为草稿 |
+| `opencli douyin publish` | 定时发布视频到抖音 |
+| `opencli douyin update` | 更新视频信息 |
+| `opencli douyin delete` | 删除作品 |
+| `opencli douyin stats` | 查询作品数据分析 |
+| `opencli douyin collections` | 获取合集列表 |
+| `opencli douyin activities` | 获取官方活动列表 |
+| `opencli douyin location` | 搜索发布可用的地理位置 |
+| `opencli douyin hashtag search` | 按关键词搜索话题 |
+| `opencli douyin hashtag suggest` | 基于封面 URI 推荐话题 |
+| `opencli douyin hashtag hot` | 获取热点词 |
+
+## Usage Examples
+
+```bash
+# 账号与作品
+opencli douyin profile
+opencli douyin videos --limit 10
+opencli douyin videos --status scheduled
+opencli douyin drafts
+
+# 发布前辅助信息
+opencli douyin collections
+opencli douyin activities
+opencli douyin location "东京塔"
+opencli douyin hashtag search "春游"
+opencli douyin hashtag hot --limit 10
+
+# 保存草稿
+opencli douyin draft ./video.mp4 \
+  --title "春游 vlog" \
+  --caption "#春游 先存草稿"
+
+# 定时发布
+opencli douyin publish ./video.mp4 \
+  --title "春游 vlog" \
+  --caption "#春游 今天去看樱花" \
+  --schedule "2026-04-08T12:00:00+09:00"
+
+# 也支持 Unix 秒字符串
+opencli douyin publish ./video.mp4 \
+  --title "春游 vlog" \
+  --schedule 1775617200
+
+# 更新与删除
+opencli douyin update 1234567890 --caption "更新后的文案"
+opencli douyin update 1234567890 --reschedule "2026-04-09T20:00:00+09:00"
+opencli douyin delete 1234567890
+
+# JSON 输出
+opencli douyin profile -f json
+```
+
+## Prerequisites
+
+- Chrome running and **logged into** `creator.douyin.com`
+- The logged-in account must have access to Douyin Creator Center publishing features
+- [Browser Bridge extension](/guide/browser-bridge) installed
+
+## Notes
+
+- `publish` requires `--schedule` to be at least 2 hours later and no more than 14 days later
+- `draft` and `publish` upload the video through Douyin/ByteDance browser-authenticated APIs, so cookies in the active browser session must be valid
+- `hashtag suggest` expects a valid `cover`/`cover_uri` value produced during the publish pipeline; for normal manual use, `hashtag search` and `hashtag hot` are usually more convenient

package/docs/adapters/browser/imdb.md

@@ -0,0 +1,47 @@
+# IMDb
+
+**Mode**: 🌐 Public (Browser) · **Domain**: `www.imdb.com`
+
+## Commands
+
+| Command | Description |
+|---------|-------------|
+| `opencli imdb search` | Search movies, TV shows, and people |
+| `opencli imdb title` | Get movie or TV show details |
+| `opencli imdb top` | IMDb Top 250 Movies |
+| `opencli imdb trending` | IMDb Most Popular Movies |
+| `opencli imdb person` | Get actor or director info |
+| `opencli imdb reviews` | Get user reviews for a title |
+
+## Usage Examples
+
+```bash
+# Search for a movie
+opencli imdb search "inception" --limit 10
+
+# Get movie details
+opencli imdb title tt1375666
+
+# Get TV series details (also accepts full URL)
+opencli imdb title "https://www.imdb.com/title/tt0903747/"
+
+# Top 250 movies
+opencli imdb top --limit 20
+
+# Currently trending movies
+opencli imdb trending --limit 10
+
+# Actor/director info with filmography
+opencli imdb person nm0634240 --limit 5
+
+# User reviews
+opencli imdb reviews tt1375666 --limit 5
+
+# JSON output
+opencli imdb top --limit 5 -f json
+```
+
+## Prerequisites
+
+- Chrome with Browser Bridge extension installed
+- No login required (all data is public)

package/docs/adapters/browser/jd.md

@@ -6,7 +6,7 @@
 
 | Command | Description |
 |---------|-------------|
-| `opencli jd item <sku>` | Fetch product details (price, images, specs) |
+| `opencli jd item <sku>` | Fetch product details (price, shop, specs, AVIF images) |
 
 ## Usage Examples
 
@@ -14,7 +14,7 @@
 # Get product details by SKU
 opencli jd item 100291143898
 
-# Limit detail images
+# Limit returned AVIF images
 opencli jd item 100291143898 --images 5
 
 # JSON output

package/docs/adapters/browser/linux-do.md

@@ -6,37 +6,198 @@
 
 | Command | Description |
 |---------|-------------|
-| `opencli linux-do hot` | 热门话题 |
-| `opencli linux-do latest` | 最新话题 |
-| `opencli linux-do categories` | 板块列表 |
-| `opencli linux-do category` | 板块话题 |
-| `opencli linux-do search` | 搜索话题 |
-| `opencli linux-do topic` | 话题详情 |
+| `opencli linux-do feed` | Browse topics (site-wide, by tag, or by category) |
+| `opencli linux-do categories` | List all categories |
+| `opencli linux-do tags` | List popular tags |
+| `opencli linux-do search <query>` | Search topics |
+| `opencli linux-do topic <id>` | View topic posts |
+| `opencli linux-do user-topics <username>` | Topics created by a user |
+| `opencli linux-do user-posts <username>` | Replies posted by a user |
 
-## Usage Examples
+## feed
+
+Browse topic listings. Defaults to latest topics when called with no arguments.
+
+- Supports filtering by `--tag`, `--category`, or both
+- `--tag` accepts tag name, slug, or ID
+- `--category` accepts category name, slug, ID, or `Parent / Child` path for sub-categories
+- Use `--view` to switch between latest / hot / top
+
+### Basic
 
 ```bash
-# Hot topics this week
-opencli linux-do hot --limit 20
+# Latest topics (default)
+opencli linux-do feed
+
+# Hot topics
+opencli linux-do feed --view hot
+
+# Top topics — default period is weekly
+opencli linux-do feed --view top
+opencli linux-do feed --view top --period daily
+opencli linux-do feed --view top --period monthly
 
-# Hot topics by period
-opencli linux-do hot --period daily
-opencli linux-do hot --period monthly
+# Sort by views descending
+opencli linux-do feed --order views
+
+# Sort by created time ascending
+opencli linux-do feed --order created --ascending
+
+# Limit results
+opencli linux-do feed --limit 10
+
+# JSON output
+opencli linux-do feed -f json
+```
 
-# Latest topics
-opencli linux-do latest --limit 10
+### Filter by tag
+
+```bash
+# By tag name, slug, or ID — all equivalent
+opencli linux-do feed --tag "ChatGPT"
+opencli linux-do feed --tag chatgpt
+opencli linux-do feed --tag 3
+
+# Tag + hot view
+opencli linux-do feed --tag "ChatGPT" --view hot
+
+# Tag + top view with period
+opencli linux-do feed --tag "OpenAI" --view top --period monthly
+```
+
+### Filter by category
+
+Supports both top-level and sub-categories. Sub-categories auto-resolve their parent path.
+
+```bash
+# Top-level category — name, slug, or ID
+opencli linux-do feed --category "开发调优"
+opencli linux-do feed --category develop
+opencli linux-do feed --category 4
 
-# List all categories
+# Sub-category
+opencli linux-do feed --category "开发调优 / Lv1"
+opencli linux-do feed --category "网盘资源"
+
+# Category + hot / top view
+opencli linux-do feed --category "开发调优" --view hot
+opencli linux-do feed --category "开发调优" --view top --period weekly
+```
+
+### Category + tag
+
+Combine `--category` and `--tag` to narrow results within a category.
+
+```bash
+opencli linux-do feed --category "开发调优" --tag "ChatGPT"
+opencli linux-do feed --category "网盘资源" --tag "OpenAI"
+opencli linux-do feed --category 94 --tag 4 --view top --period monthly
+```
+
+### Parameters
+
+| Parameter | Description | Default |
+|-----------|-------------|---------|
+| `--view V` | `latest`, `hot`, `top` | `latest` |
+| `--tag VALUE` | Tag name, slug, or ID | — |
+| `--category VALUE` | Category name, slug, or ID | — |
+| `--limit N` | Number of results | `20` |
+| `--order O` | `default`, `created`, `activity`, `views`, `posts`, `category`, `likes`, `op_likes`, `posters` | `default` |
+| `--ascending` | Sort ascending instead of descending | off |
+| `--period P` | `all`, `daily`, `weekly`, `monthly`, `quarterly`, `yearly` (only with `--view top`) | `weekly` |
+
+Output columns: `title`, `replies`, `created`, `likes`, `views`, `url`
+
+## categories
+
+List forum categories with optional sub-category expansion.
+
+```bash
 opencli linux-do categories
+opencli linux-do categories --subcategories
+opencli linux-do categories --limit 50
+```
+
+When `--subcategories` is enabled, sub-categories are rendered as `Parent / Child` so the `name` value can be copied directly into `opencli linux-do feed --category ...`.
+
+Output columns: `name`, `slug`, `id`, `topics`, `description`
+
+## tags
 
-# Search topics
+List tags sorted by usage count.
+
+```bash
+opencli linux-do tags
+opencli linux-do tags --limit 50
+```
+
+Output columns: `rank`, `name`, `count`, `url`
+
+## search
+
+Search topics by keyword.
+
+```bash
 opencli linux-do search "NixOS"
+opencli linux-do search "Docker" --limit 10
+opencli linux-do search "Claude" -f json
+```
 
-# View topic details
-opencli linux-do topic 12345
+Output columns: `rank`, `title`, `views`, `likes`, `replies`, `url`
 
-# JSON output
-opencli linux-do hot -f json
+## topic
+
+View posts within a topic (first page).
+
+```bash
+opencli linux-do topic 1234
+opencli linux-do topic 1234 --limit 50
+opencli linux-do topic 1234 --main_only -f json | jq -r '.[0].content'
+```
+
+Notes:
+- `--main_only` returns only the main post row and keeps the body untruncated
+
+Output columns: `author`, `content`, `likes`, `created_at`
+
+## user-topics
+
+List topics created by a user.
+
+```bash
+opencli linux-do user-topics neo
+opencli linux-do user-topics neo --limit 10
+```
+
+Output columns: `rank`, `title`, `replies`, `created_at`, `likes`, `views`, `url`
+
+## user-posts
+
+List replies posted by a user.
+
+```bash
+opencli linux-do user-posts neo
+opencli linux-do user-posts neo --limit 10
+```
+
+Output columns: `index`, `topic_user`, `topic`, `reply`, `time`, `url`
+
+## Compatibility
+
+The legacy commands below are still available as compatibility wrappers while `feed` becomes the canonical entrypoint:
+
+```bash
+opencli linux-do latest
+opencli linux-do hot --period weekly
+opencli linux-do category develop 4
+```
+
+Preferred modern forms:
+
+```bash
+opencli linux-do feed --view latest
+opencli linux-do feed --view top --period weekly
+opencli linux-do feed --category 4
 ```
 
 ## Prerequisites

package/docs/adapters/browser/paperreview.md

@@ -0,0 +1,43 @@
+# paperreview.ai
+
+**Mode**: 🌐 Public · **Domain**: `paperreview.ai`
+
+## Commands
+
+| Command | Description |
+|---------|-------------|
+| `opencli paperreview submit` | Submit a PDF to paperreview.ai for review |
+| `opencli paperreview review` | Fetch a review by token |
+| `opencli paperreview feedback` | Send feedback on a completed review |
+
+## Usage Examples
+
+```bash
+# Validate a local PDF without uploading it
+opencli paperreview submit ./paper.pdf --email you@example.com --venue RAL --dry-run true
+
+# Request an upload slot but stop before the actual upload
+opencli paperreview submit ./paper.pdf --email you@example.com --venue RAL --prepare-only true
+
+# Submit a paper for review
+opencli paperreview submit ./paper.pdf --email you@example.com --venue RAL -f json
+
+# Check the review status or fetch the final review
+opencli paperreview review tok_123 -f json
+
+# Submit feedback on the review quality
+opencli paperreview feedback tok_123 --helpfulness 4 --critical-error no --actionable-suggestions yes
+```
+
+## Prerequisites
+
+- No browser required — uses public paperreview.ai endpoints
+- The input file must be a local `.pdf`
+- paperreview.ai currently rejects files larger than `10MB`
+- `submit` requires `--email`; `--venue` is optional
+
+## Notes
+
+- `submit` returns both the review token and the review URL when submission succeeds
+- `review` returns `processing` until the paperreview.ai result is ready
+- `feedback` expects `yes` / `no` values for `--critical-error` and `--actionable-suggestions`

package/docs/adapters/browser/producthunt.md

@@ -0,0 +1,49 @@
+# Product Hunt
+
+**Mode**: 🌐 Public / 🔐 Browser · **Domain**: `www.producthunt.com`
+
+## Commands
+
+| Command | Description |
+|---------|-------------|
+| `opencli producthunt posts` | Latest Product Hunt launches (optional category filter) |
+| `opencli producthunt today` | Today's Product Hunt launches (most recent day in feed) |
+| `opencli producthunt hot` | Today's top Product Hunt launches with vote counts |
+| `opencli producthunt browse <category>` | Best products in a Product Hunt category |
+
+## Usage Examples
+
+```bash
+# Today's top launches with vote counts
+opencli producthunt hot --limit 10
+
+# Latest posts (RSS feed)
+opencli producthunt posts --limit 20
+
+# Filter by category
+opencli producthunt posts --category developer-tools --limit 10
+
+# Today's launches only
+opencli producthunt today --limit 10
+
+# Browse best products in a category
+opencli producthunt browse vibe-coding --limit 10
+opencli producthunt browse ai-agents --limit 10
+opencli producthunt browse developer-tools --limit 10
+
+# JSON output
+opencli producthunt hot -f json
+```
+
+## Category Slugs
+
+Common categories for `browse` and `posts --category`:
+
+`ai-agents`, `ai-coding-agents`, `ai-code-editors`, `ai-chatbots`, `ai-workflow-automation`,
+`vibe-coding`, `developer-tools`, `productivity`, `design-creative`, `marketing-sales`,
+`no-code-platforms`, `llms`, `finance`, `social-community`, `engineering-development`
+
+## Prerequisites
+
+- `posts` and `today` — no browser required (public RSS feed)
+- `hot` and `browse` — Chrome running with [Browser Bridge extension](/guide/browser-bridge) installed

package/docs/adapters/browser/twitter.md

@@ -37,6 +37,12 @@
 # Quick start
 opencli twitter trending --limit 5
 
+# Search top tweets (default)
+opencli twitter search "react 19"
+
+# Search latest/live tweets
+opencli twitter search "react 19" --filter live
+
 # JSON output
 opencli twitter trending -f json
 

package/docs/adapters/desktop/chatgpt.md

@@ -14,8 +14,13 @@ The current built-in commands use native AppleScript automation — no extra lau
 - `opencli chatgpt status`: Check if the ChatGPT app is currently running.
 - `opencli chatgpt new`: Activate ChatGPT and press `Cmd+N` to start a new conversation.
 - `opencli chatgpt send "message"`: Copy your message to clipboard, activate ChatGPT, paste, and submit.
+- `opencli chatgpt send "message" --model thinking`: Switch model/mode first, then send the message.
 - `opencli chatgpt read`: Read the last visible message from the focused ChatGPT window via the Accessibility tree.
 - `opencli chatgpt ask "message"`: Send a prompt and wait for the visible reply in one shot.
+- `opencli chatgpt ask "message" --model instant`: Run a one-shot prompt using a specific model/mode.
+- `opencli chatgpt model thinking`: Switch the active ChatGPT model/mode without sending a message.
+
+Supported model choices: `auto`, `instant`, `thinking`, `5.2-instant`, `5.2-thinking`.
 
 ## Approach 2: CDP (Advanced, Electron Debug Mode)
 

package/docs/adapters/index.md

@@ -11,7 +11,7 @@ Run `opencli list` for the live registry.
 | **[bilibili](/adapters/browser/bilibili)** | `hot` `search` `me` `favorite` `history` `feed` `subtitle` `dynamic` `ranking` `following` `user-videos` `download` | 🔐 Browser |
 | **[zhihu](/adapters/browser/zhihu)** | `hot` `search` `question` `download` | 🔐 Browser |
 | **[xiaohongshu](/adapters/browser/xiaohongshu)** | `search` `notifications` `feed` `user` `download` `publish` `creator-notes` `creator-note-detail` `creator-notes-summary` `creator-profile` `creator-stats` | 🔐 Browser |
-| **[xueqiu](/adapters/browser/xueqiu)** | `feed` `hot-stock` `hot` `search` `stock` `watchlist` `earnings-date` | 🔐 Browser |
+| **[xueqiu](/adapters/browser/xueqiu)** | `feed` `hot-stock` `hot` `search` `stock` `watchlist` `earnings-date` `fund-holdings` `fund-snapshot` | 🔐 Browser |
 | **[youtube](/adapters/browser/youtube)** | `search` `video` `transcript` | 🔐 Browser |
 | **[v2ex](/adapters/browser/v2ex)** | `hot` `latest` `topic` `node` `user` `member` `replies` `nodes` `daily` `me` `notifications` | 🌐 / 🔐 |
 | **[bloomberg](/adapters/browser/bloomberg)** | `main` `markets` `economics` `industries` `tech` `politics` `businessweek` `opinions` `feeds` `news` | 🌐 / 🔐 |
@@ -25,19 +25,26 @@ Run `opencli list` for the live registry.
 | **[jike](/adapters/browser/jike)** | `feed` `search` `post` `topic` `user` `create` `comment` `like` `repost` `notifications` | 🔐 Browser |
 | **[jimeng](/adapters/browser/jimeng)** | `generate` `history` | 🔐 Browser |
 | **[yollomi](/adapters/browser/yollomi)** | `generate` `video` `edit` `upload` `models` `remove-bg` `upscale` `face-swap` `restore` `try-on` `background` `object-remover` | 🔐 Browser |
-| **[linux-do](/adapters/browser/linux-do)** | `hot` `latest` `categories` `category` `search` `topic` | 🔐 Browser |
+| **[linux-do](/adapters/browser/linux-do)** | `feed` `categories` `tags` `search` `topic` `user-topics` `user-posts` | 🔐 Browser |
 | **[chaoxing](/adapters/browser/chaoxing)** | `assignments` `exams` | 🔐 Browser |
 | **[grok](/adapters/browser/grok)** | `ask` | 🔐 Browser |
 | **[doubao](/adapters/browser/doubao)** | `status` `new` `send` `read` `ask` | 🔐 Browser |
 | **[weread](/adapters/browser/weread)** | `shelf` `search` `book` `ranking` `notebooks` `highlights` `notes` | 🔐 Browser |
-| **[douban](/adapters/browser/douban)** | `search` `top250` `subject` `marks` `reviews` `movie-hot` `book-hot` | 🔐 Browser |
+| **[douban](/adapters/browser/douban)** | `search` `top250` `subject` `photos` `download` `marks` `reviews` `movie-hot` `book-hot` | 🔐 Browser |
 | **[facebook](/adapters/browser/facebook)** | `feed` `profile` `search` `friends` `groups` `events` `notifications` `memories` `add-friend` `join-group` | 🔐 Browser |
+| **[imdb](/adapters/browser/imdb)** | `search` `title` `top` `trending` `person` `reviews` | 🌐 / 🔐 |
 | **[instagram](/adapters/browser/instagram)** | `explore` `profile` `search` `user` `followers` `following` `follow` `unfollow` `like` `unlike` `comment` `save` `unsave` `saved` | 🔐 Browser |
 | **[medium](/adapters/browser/medium)** | `feed` `search` `user` | 🔐 Browser |
 | **[sinablog](/adapters/browser/sinablog)** | `hot` `search` `article` `user` | 🔐 Browser |
 | **[substack](/adapters/browser/substack)** | `feed` `search` `publication` | 🔐 Browser |
 | **[pixiv](/adapters/browser/pixiv)** | `ranking` `search` `user` `illusts` `detail` `download` | 🔐 Browser |
 | **[tiktok](/adapters/browser/tiktok)** | `explore` `search` `profile` `user` `following` `follow` `unfollow` `like` `unlike` `comment` `save` `unsave` `live` `notifications` `friends` | 🔐 Browser |
+| **[google](/adapters/browser/google)** | `news` `search` `suggest` `trends` | 🌐 / 🔐 |
+| **[jd](/adapters/browser/jd)** | `item` | 🔐 Browser |
+| **[web](/adapters/browser/web)** | `read` | 🔐 Browser |
+| **[weixin](/adapters/browser/weixin)** | `download` | 🔐 Browser |
+| **[36kr](/adapters/browser/36kr)** | `news` `hot` `search` `article` | 🌐 / 🔐 |
+| **[producthunt](/adapters/browser/producthunt)** | `posts` `today` `hot` `browse` | 🌐 / 🔐 |
 
 ## Public API Adapters
 
@@ -51,12 +58,14 @@ Run `opencli list` for the live registry.
 | **[xiaoyuzhou](/adapters/browser/xiaoyuzhou)** | `podcast` `podcast-episodes` `episode` | 🌐 Public |
 | **[yahoo-finance](/adapters/browser/yahoo-finance)** | `quote` | 🌐 Public |
 | **[arxiv](/adapters/browser/arxiv)** | `search` `paper` | 🌐 Public |
+| **[paperreview](/adapters/browser/paperreview)** | `submit` `review` `feedback` | 🌐 Public |
 | **[barchart](/adapters/browser/barchart)** | `quote` `options` `greeks` `flow` | 🌐 Public |
 | **[hf](/adapters/browser/hf)** | `top` | 🌐 Public |
 | **[sinafinance](/adapters/browser/sinafinance)** | `news` | 🌐 Public |
 | **[stackoverflow](/adapters/browser/stackoverflow)** | `hot` `search` `bounties` `unanswered` | 🌐 Public |
 | **[wikipedia](/adapters/browser/wikipedia)** | `search` `summary` `random` `trending` | 🌐 Public |
 | **[lobsters](/adapters/browser/lobsters)** | `hot` `newest` `active` `tag` | 🌐 Public |
+| **[steam](/adapters/browser/steam)** | `top-sellers` | 🌐 Public |
 
 ## Desktop Adapters
 

package/docs/advanced/download.md

@@ -9,6 +9,7 @@ OpenCLI supports downloading images, videos, and articles from supported platfor
 | **xiaohongshu** | Images, Videos | Downloads all media from a note |
 | **bilibili** | Videos | Requires `yt-dlp` installed |
 | **twitter** | Images, Videos | Downloads from user media tab or single tweet |
+| **douban** | Images | Downloads poster / still image lists from movie subjects |
 | **zhihu** | Articles (Markdown) | Exports articles with optional image download |
 | **weixin** | Articles (Markdown) | Exports WeChat Official Account articles |
 
@@ -39,6 +40,9 @@ opencli twitter download elonmusk --limit 20 --output ./twitter
 # Download single tweet media
 opencli twitter download --tweet-url "https://x.com/user/status/123" --output ./twitter
 
+# Download Douban posters / stills
+opencli douban download 30382501 --output ./douban
+
 # Export Zhihu article to Markdown
 opencli zhihu download "https://zhuanlan.zhihu.com/p/xxx" --output ./zhihu
 

package/docs/advanced/rate-limiter-plugin.md

@@ -0,0 +1,99 @@
+# Rate Limiter Plugin
+
+An optional plugin that adds a random sleep between browser-based commands to reduce the risk of platform rate-limiting or bot detection.
+
+## Install
+
+```bash
+opencli plugin install github:jackwener/opencli-plugin-rate-limiter
+```
+
+Or copy the example below into `~/.opencli/plugins/rate-limiter/` to use it locally without installing from GitHub.
+
+## What it does
+
+After every command targeting a browser platform (xiaohongshu, weibo, bilibili, douyin, tiktok, …), the plugin sleeps for a random duration — 5–30 seconds by default — before returning control to the caller.
+
+## Configuration
+
+| Variable | Default | Description |
+|---|---|---|
+| `OPENCLI_RATE_MIN` | `5` | Minimum sleep in seconds |
+| `OPENCLI_RATE_MAX` | `30` | Maximum sleep in seconds |
+| `OPENCLI_NO_RATE` | — | Set to `1` to disable entirely (local dev) |
+
+```bash
+# Shorter delays for light scraping
+OPENCLI_RATE_MIN=3 OPENCLI_RATE_MAX=10 opencli xiaohongshu search "AI眼镜"
+
+# Skip delays when iterating locally
+OPENCLI_NO_RATE=1 opencli bilibili comments BV1WtAGzYEBm
+```
+
+## Local installation (without GitHub)
+
+1. Create the plugin directory:
+
+   ```bash
+   mkdir -p ~/.opencli/plugins/rate-limiter
+   ```
+
+2. Create `~/.opencli/plugins/rate-limiter/package.json`:
+
+   ```json
+   { "type": "module" }
+   ```
+
+3. Create `~/.opencli/plugins/rate-limiter/index.js`:
+
+   ```js
+   import { onAfterExecute } from '@jackwener/opencli/hooks'
+
+   const BROWSER_DOMAINS = [
+     'xiaohongshu', 'weibo', 'bilibili', 'douyin', 'tiktok',
+     'instagram', 'twitter', 'youtube', 'zhihu', 'douban',
+     'jike', 'weixin', 'xiaoyuzhou',
+   ]
+
+   onAfterExecute(async (ctx) => {
+     if (process.env.OPENCLI_NO_RATE === '1') return
+
+     const site = ctx.command?.split('/')?.[0] ?? ''
+     if (!BROWSER_DOMAINS.includes(site)) return
+
+     const min = Number(process.env.OPENCLI_RATE_MIN ?? 5)
+     const max = Number(process.env.OPENCLI_RATE_MAX ?? 30)
+     const ms = Math.floor(Math.random() * (max - min + 1) + min) * 1000
+
+     process.stderr.write(`[rate-limiter] ${site}: sleeping ${(ms / 1000).toFixed(0)}s\n`)
+     await new Promise(r => setTimeout(r, ms))
+   })
+   ```
+
+4. Verify it loaded:
+
+   ```bash
+   OPENCLI_NO_RATE=1 opencli xiaohongshu search "test" 2>&1 | grep rate-limiter
+   # → (no output — plugin loaded but rate limit skipped)
+
+   opencli xiaohongshu search "test" 2>&1 | grep rate-limiter
+   # → [rate-limiter] xiaohongshu: sleeping 12s
+   ```
+
+## Writing your own plugin
+
+Plugins are plain JS/TS files in `~/.opencli/plugins/<name>/`. A plugin file must export a hook registration call that matches the pattern `onStartup(`, `onBeforeExecute(`, or `onAfterExecute(` — opencli's discovery engine uses this pattern to identify hook files vs. command files.
+
+```js
+// ~/.opencli/plugins/my-plugin/index.js
+import { onAfterExecute } from '@jackwener/opencli/hooks'
+
+onAfterExecute(async (ctx) => {
+  // ctx.command — e.g. "bilibili/comments"
+  // ctx.args    — coerced command arguments
+  // ctx.error   — set if the command threw
+  console.error(`[my-plugin] finished: ${ctx.command}`)
+})
+```
+
+See [hooks.ts](../../src/hooks.ts) for the full `HookContext` type.