12bd727e30
问题：bird CLI 用 Node.js 原生 fetch()，不走 HTTP_PROXY， 导致需要代理的用户（如 Windows + 本地代理）完全无法使用。 修复： 1. 自动代理注入：检测到 HTTP_PROXY/HTTPS_PROXY 时，通过 undici 的 EnvHttpProxyAgent 让 Node.js fetch 自动走代理，用户无需任何额外配置 2. doctor 检测：实际跑 bird whoami 测连通性，失败时给出具体原因 3. search fallback：bird 搜索失败时自动 fallback 到 Exa 4. install：自动安装 undici（npm install -g undici） 5. 新增 troubleshooting.md：常见问题解决方案 Fixes #9 |
||
|---|---|---|
| .. | ||
| assets | ||
| cookie-export.md | ||
| install.md | ||
| README_en.md | ||
| troubleshooting.md | ||
👁️ Agent Reach
Give your AI Agent one-click access to the entire internet
Quick Start · 中文 · Platforms · Philosophy
Why Agent Reach?
AI Agents can already access the internet — but "can go online" is barely the start.
The most valuable information lives across social and niche platforms: Twitter discussions, Reddit feedback, YouTube tutorials, XiaoHongShu reviews, Bilibili videos, GitHub activity… These are where information density is highest, but each platform has its own barriers:
| Pain Point | Reality |
|---|---|
| Twitter API | Pay-per-use, moderate usage ~$215/month |
| Server IPs get 403'd | |
| XiaoHongShu | Login required to browse |
| Bilibili | Blocks overseas/server IPs |
To connect your Agent to these platforms, you'd have to find tools, install dependencies, and debug configs — one by one.
Agent Reach turns this into one command:
Install Agent Reach: https://raw.githubusercontent.com/Panniantong/agent-reach/main/docs/install.md
Copy that to your Agent. A few minutes later, it can read tweets, search Reddit, and watch Bilibili.
✅ Before you start, you might want to know
| 💰 Completely free | All tools are open source, all APIs are free. The only possible cost is a server proxy ($1/month) — local computers don't need one |
| 🔒 Privacy safe | Cookies stay local. Never uploaded. Fully open source — audit anytime |
| 🔄 Kept up to date | Upstream tools (yt-dlp, bird, Jina Reader, etc.) are tracked and updated regularly |
| 🤖 Works with any Agent | Claude Code, OpenClaw, Cursor, Windsurf… any Agent that can run commands |
| 🩺 Built-in diagnostics | agent-reach doctor — one command shows what works, what doesn't, and how to fix it |
Supported Platforms
| Platform | Capabilities | Setup | Notes |
|---|---|---|---|
| 🌐 Web | Read | Zero config | Any URL → clean Markdown (Jina Reader ⭐9.8K) |
| 🐦 Twitter/X | Read · Search | Zero config / Cookie | Single tweets readable out of the box. Cookie unlocks search, timeline, posting (bird) |
| 📕 XiaoHongShu | Read · Search · Post · Comment · Like | mcporter | Via xiaohongshu-mcp internal API, install and go |
| Search (via Exa) | Read posts and profiles | Tell your Agent "help me set up Instagram" | |
| Jina Reader (public pages) | Full profiles, companies, job search | Tell your Agent "help me set up LinkedIn" | |
| 🏢 Boss直聘 | Jina Reader (job pages) | Job search, greet recruiters | Tell your Agent "help me set up Boss直聘" |
| 🔍 Web Search | Search | Auto-configured | Auto-configured during install, free, no API key (Exa via mcporter) |
| 📦 GitHub | Read · Search | Zero config | gh CLI powered. Public repos work immediately. gh auth login unlocks Fork, Issue, PR |
| 📺 YouTube | Read · Search | Zero config | Subtitles + search across 1800+ video sites (yt-dlp ⭐148K) |
| 📺 Bilibili | Read · Search | Zero config / Proxy | Video info + subtitles + search. Local works directly, servers need a proxy (yt-dlp) |
| 📡 RSS | Read | Zero config | Any RSS/Atom feed (feedparser ⭐2.3K) |
| Search · Read | Free / Proxy | Search via Exa (free). Reading posts needs a proxy on servers |
Setup levels: Zero config = install and go · Auto-configured = handled during install · mcporter = needs MCP service · Cookie = export from browser · Proxy = $1/month
Quick Start
Copy this to your AI Agent (Claude Code, OpenClaw, Cursor, etc.):
Install Agent Reach: https://raw.githubusercontent.com/Panniantong/agent-reach/main/docs/install.md
The Agent auto-installs, detects your environment, and tells you what's ready.
Manual install
pip install https://github.com/Panniantong/agent-reach/archive/main.zip
agent-reach install --env=auto
Install as a Skill (Claude Code / OpenClaw / any agent with Skills support)
npx skills add Panniantong/Agent-Reach@agent-reach
After the Skill is installed, the Agent will auto-detect whether agent-reach CLI is available and install it if needed.
If you install via
agent-reach install, the skill is registered automatically — no extra steps needed.
Works Out of the Box
No configuration needed — just tell your Agent:
- "Read this link" → any web page
- "What's this GitHub repo about?" → repos, issues, code
- "What does this video cover?" → YouTube / Bilibili subtitles
- "Read this tweet" → Twitter posts
- "Subscribe to this RSS" → RSS / Atom feeds
- "Search GitHub for LLM frameworks" → GitHub search
No commands to remember. The Agent knows what to call.
Unlock on Demand
Don't use it? Don't configure it. Every step is optional.
🍪 Cookies — Free, 2 minutes
Tell your Agent "help me configure Twitter cookies" — it'll guide you through exporting from your browser. Local computers can auto-import.
🌐 Proxy — $1/month, servers only
Reddit and Bilibili block server IPs. Get a proxy (Webshare recommended, $1/month) and send the address to your Agent.
Local computers don't need a proxy. Reddit search works free via Exa even without one.
Status at a Glance
$ agent-reach doctor
👁️ Agent Reach Status
========================================
✅ Ready to use:
✅ GitHub repos and code — public repos readable and searchable
✅ Twitter/X tweets — readable. Cookie unlocks search and posting
✅ YouTube video subtitles — yt-dlp
⚠️ Bilibili video info — server IPs may be blocked, configure proxy
✅ RSS/Atom feeds — feedparser
✅ Web pages (any URL) — Jina Reader API
🔍 Search (free Exa key to unlock):
⬜ Web semantic search — sign up at exa.ai for free key
🔧 Configurable:
⬜ Reddit posts and comments — search via Exa (free). Reading needs proxy
⬜ XiaoHongShu notes — needs cookie. Export from browser
Status: 6/9 channels available
Design Philosophy
Agent Reach is a scaffolding tool, not a framework.
Every time you spin up a new Agent, you spend time finding tools, installing deps, and debugging configs — what reads Twitter? How do you bypass Reddit blocks? How do you extract YouTube subtitles? Every time, you re-do the same work.
Agent Reach does one simple thing: it makes those tool selection and configuration decisions for you.
🔌 Every Channel is Pluggable
Each platform is a single Python file implementing a unified interface. Backends can be swapped anytime — when a better tool comes along, change one file and nothing else breaks.
channels/
├── web.py → Jina Reader ← swap to Firecrawl, Crawl4AI…
├── twitter.py → bird ← swap to Nitter, official API…
├── youtube.py → yt-dlp ← swap to YouTube API, Whisper…
├── github.py → gh CLI ← swap to REST API, PyGithub…
├── bilibili.py → yt-dlp ← swap to bilibili-api…
├── reddit.py → JSON API + Exa ← swap to PRAW, Pushshift…
├── xiaohongshu.py → mcporter MCP ← swap to other XHS tools…
├── instagram.py → instaloader ← swap to instagrapi, official API…
├── linkedin.py → linkedin-mcp ← swap to LinkedIn API…
├── bosszhipin.py → mcp-bosszp ← swap to other job tools…
├── rss.py → feedparser ← swap to atoma…
├── exa_search.py → mcporter MCP ← swap to Tavily, SerpAPI…
└── __init__.py → Channel registry
Current Tool Choices
| Scenario | Tool | Why |
|---|---|---|
| Read web pages | Jina Reader | 9.8K stars, free, no API key needed |
| Read tweets | bird | Cookie auth, free. Official API is pay-per-use ($0.005/post read) |
| Video subtitles + search | yt-dlp | 148K stars, YouTube + Bilibili + 1800 sites |
| Search the web | Exa via mcporter | AI semantic search, MCP integration, no API key |
| GitHub | gh CLI | Official tool, full API after auth |
| Read RSS | feedparser | Python ecosystem standard, 2.3K stars |
| XiaoHongShu | xiaohongshu-mcp | Internal API, bypasses anti-bot |
| instaloader | 9.8K stars, Python CLI, cookie auth, free | |
| linkedin-scraper-mcp | 900+ stars, MCP server, browser automation | |
| Boss直聘 | mcp-bosszp | MCP server, job search + recruiter greeting |
📌 These are the current choices. Don't like one? Swap out the file. That's the whole point of scaffolding.
Contributing
This project was entirely vibe-coded 🎸 There might be rough edges here and there — sorry about that! If you run into any bugs, please don't hesitate to open an Issue and I'll fix it ASAP.
Want a new channel? Open an Issue to request it, or submit a PR yourself.
Want to add one locally? Just have your Agent clone the repo and modify it — each channel is a single standalone file, easy to add.
PRs always welcome!
Credits
Jina Reader · yt-dlp · bird · Exa · feedparser · instaloader · linkedin-scraper-mcp · mcp-bosszp