[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-74786":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":9,"language":10,"languages":9,"totalLinesOfCode":9,"stars":11,"forks":12,"watchers":13,"openIssues":14,"contributorsCount":15,"subscribersCount":15,"size":15,"stars1d":16,"stars7d":17,"stars30d":18,"stars90d":15,"forks30d":15,"starsTrendScore":19,"compositeScore":20,"rankGlobal":9,"rankLanguage":9,"license":21,"archived":22,"fork":22,"defaultBranch":23,"hasWiki":24,"hasPages":22,"topics":25,"createdAt":9,"pushedAt":9,"updatedAt":26,"readmeContent":27,"aiSummary":28,"trendingCount":15,"starSnapshotCount":15,"syncStatus":29,"lastSyncTime":30,"discoverSource":31},74786,"bb-browser","epiral\u002Fbb-browser","epiral","Your browser is the API. CLI + MCP server for AI agents to control Chrome with your login state.",null,"TypeScript",5735,568,13,63,0,45,116,648,135,39.27,"MIT License",false,"main",true,[],"2026-06-12 02:03:28","\u003Cdiv align=\"center\">\n\n# bb-browser\n\n### BadBoy Browser\n\n**Your browser is the API. No keys. No bots. No scrapers.**\n\n[![npm](https:\u002F\u002Fimg.shields.io\u002Fnpm\u002Fv\u002Fbb-browser?color=CB3837&logo=npm&logoColor=white)](https:\u002F\u002Fwww.npmjs.com\u002Fpackage\u002Fbb-browser)\n[![Node.js](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FNode.js-18+-339933?logo=node.js&logoColor=white)](https:\u002F\u002Fnodejs.org)\n[![License](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-MIT-blue.svg)](LICENSE)\n\n[English](README.md) · [中文](README.zh-CN.md)\n\n\u003C\u002Fdiv>\n\n---\n\nYou're already logged into Twitter, Reddit, YouTube, Zhihu, Bilibili, LinkedIn, GitHub — bb-browser lets AI agents **use that directly**.\n\n```bash\nbb-browser site twitter\u002Fsearch \"AI agent\"       # search tweets\nbb-browser site zhihu\u002Fhot                        # trending on Zhihu\nbb-browser site arxiv\u002Fsearch \"transformer\"       # search papers\nbb-browser site eastmoney\u002Fstock \"茅台\"            # real-time stock quote\nbb-browser site boss\u002Fsearch \"AI engineer\"        # search jobs\nbb-browser site wikipedia\u002Fsummary \"Python\"       # Wikipedia summary\nbb-browser site youtube\u002Ftranscript VIDEO_ID      # full transcript\nbb-browser site stackoverflow\u002Fsearch \"async\"     # search SO questions\n```\n\n**103 commands across 36 platforms.** All using your real browser's login state. [Full list →](https:\u002F\u002Fgithub.com\u002Fepiral\u002Fbb-sites)\n\n## The idea\n\nThe internet was built for browsers. AI agents have been trying to access it through APIs — but 99% of websites don't offer one.\n\nbb-browser flips this: **instead of forcing websites to provide machine interfaces, let machines use the human interface directly.** The adapter runs `eval` inside your browser tab, calls `fetch()` with your cookies, or invokes the page's own webpack modules. The website thinks it's you. Because it **is** you.\n\n| | Playwright \u002F Selenium | Scraping libs | bb-browser |\n|---|---|---|---|\n| Browser | Headless, isolated | No browser | Your real Chrome |\n| Login state | None, must re-login | Cookie extraction | Already there |\n| Anti-bot | Detected easily | Cat-and-mouse | Invisible — it IS the user |\n| Complex auth | Can't replicate | Reverse engineer | Page handles it itself |\n\n## Quick Start\n\n### Install\n\n```bash\nnpm install -g bb-browser\n```\n\n### Use\n\n```bash\nbb-browser site update        # pull community adapters\nbb-browser site recommend     # see which adapters match your browsing habits\nbb-browser site zhihu\u002Fhot     # go\n```\n\n### OpenClaw (no extension needed)\n\nIf you use [OpenClaw](https:\u002F\u002Fopenclaw.ai), bb-browser runs directly through OpenClaw's built-in browser — no Chrome extension or daemon required:\n\n```bash\nbb-browser site reddit\u002Fhot --openclaw\nbb-browser site xueqiu\u002Fhot-stock 5 --openclaw --jq '.items[] | {name, changePercent}'\n```\n\nSkill on ClawHub: [bb-browser-openclaw](https:\u002F\u002Fclawhub.ai\u002Fyan5xu\u002Fbb-browser)\n\n### MCP (Claude Code \u002F Cursor)\n\n```json\n{\n  \"mcpServers\": {\n    \"bb-browser\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"bb-browser\", \"--mcp\"]\n    }\n  }\n}\n```\n\n## 36 platforms, 103 commands\n\nCommunity-driven via [bb-sites](https:\u002F\u002Fgithub.com\u002Fepiral\u002Fbb-sites). One JS file per command.\n\n| Category | Platforms | Commands |\n|----------|-----------|----------|\n| **Search** | Google, Baidu, Bing, DuckDuckGo, Sogou WeChat | search |\n| **Social** | Twitter\u002FX, Reddit, Weibo, Xiaohongshu, Jike, LinkedIn, Hupu | search, feed, thread, user, notifications, hot |\n| **News** | BBC, Reuters, 36kr, Toutiao, Eastmoney | headlines, search, newsflash, hot |\n| **Dev** | GitHub, StackOverflow, HackerNews, CSDN, cnblogs, V2EX, Dev.to, npm, PyPI, arXiv | search, issues, repo, top, thread, package |\n| **Video** | YouTube, Bilibili | search, video, transcript, popular, comments, feed |\n| **Entertainment** | Douban, IMDb, Genius, Qidian | movie, search, top250 |\n| **Finance** | Xueqiu, Eastmoney, Yahoo Finance | stock, hot stocks, feed, watchlist, search |\n| **Jobs** | BOSS Zhipin, LinkedIn | search, detail, profile |\n| **Knowledge** | Wikipedia, Zhihu, Open Library | search, summary, hot, question |\n| **Shopping** | SMZDM | search deals |\n| **Tools** | Youdao, GSMArena, Product Hunt, Ctrip | translate, phone specs, trending products |\n\n## 10 minutes to add any website\n\n```bash\nbb-browser guide    # full tutorial\n```\n\nTell your AI agent: *\"turn XX website into a CLI\"*. It reads the guide, reverse-engineers the API with `network --with-body`, writes the adapter, tests it, and submits a PR. All autonomously.\n\nThree tiers of adapter complexity:\n\n| Tier | Auth method | Example | Time |\n|------|-------------|---------|------|\n| **1** | Cookie (fetch directly) | Reddit, GitHub, V2EX | ~1 min |\n| **2** | Bearer + CSRF token | Twitter, Zhihu | ~3 min |\n| **3** | Webpack injection \u002F Pinia store | Twitter search, Xiaohongshu | ~10 min |\n\nWe tested this: **20 AI agents ran in parallel, each independently reverse-engineered a website and produced a working adapter.** The marginal cost of adding a new website to the agent-accessible internet is approaching zero.\n\n## What this means for AI agents\n\nWithout bb-browser, an AI agent's world is: **files + terminal + a few APIs with keys.**\n\nWith bb-browser: **files + terminal + the entire internet.**\n\nAn agent can now, in under a minute:\n\n```bash\n# Cross-platform research on any topic\nbb-browser site arxiv\u002Fsearch \"retrieval augmented generation\"\nbb-browser site twitter\u002Fsearch \"RAG\"\nbb-browser site github search rag-framework\nbb-browser site stackoverflow\u002Fsearch \"RAG implementation\"\nbb-browser site zhihu\u002Fsearch \"RAG\"\nbb-browser site 36kr\u002Fnewsflash\n```\n\nSix platforms, six dimensions, structured JSON. Faster and broader than any human researcher.\n\n## Also a full browser automation tool\n\n```bash\nbb-browser open https:\u002F\u002Fexample.com\nbb-browser snapshot -i                # accessibility tree\nbb-browser click @3                   # click element\nbb-browser fill @5 \"hello\"            # fill input\nbb-browser eval \"document.title\"      # run JS\nbb-browser fetch URL --json           # authenticated fetch\nbb-browser network requests --with-body --json  # capture traffic\nbb-browser screenshot                 # take screenshot\n```\n\nAll commands support `--json` output, `--jq \u003Cexpr>` for inline filtering, and `--tab \u003Cid>` for concurrent multi-tab operations.\n\n```bash\nbb-browser site xueqiu\u002Fhot-stock 5 --jq '.items[] | {name, changePercent}'\n# {\"name\":\"云天化\",\"changePercent\":\"2.08%\"}\n# {\"name\":\"东芯股份\",\"changePercent\":\"-7.60%\"}\n\nbb-browser site info xueqiu\u002Fstock   # view adapter args, example, domain\n```\n\n## Daemon configuration\n\nThe daemon binds to `127.0.0.1:19824` by default. You can customize the host with `--host`:\n\n```bash\nbb-browser daemon --host 127.0.0.1    # IPv4 only (fix macOS IPv6 issues)\nbb-browser daemon --host 0.0.0.0      # listen on all interfaces (for Tailscale \u002F ZeroTier remote access)\n```\n\n## Architecture\n\n```\nAI Agent (Claude Code, Codex, Cursor, etc.)\n       │ CLI or MCP (stdio)\n       ▼\nbb-browser CLI ──HTTP──▶ Daemon ──CDP WebSocket──▶ Your Real Browser\n                           │\n                    ┌──────┴──────┐\n                    │ Per-tab     │\n                    │ event cache │\n                    │ (network,   │\n                    │  console,   │\n                    │  errors)    │\n                    └─────────────┘\n```\n\n## License\n\n[MIT](LICENSE)\n","bb-browser 是一个让AI代理能够直接利用你浏览器登录状态访问网络的工具。它通过命令行界面和MCP服务器控制Chrome，支持103个命令横跨36个平台，如Twitter、Reddit、YouTube等。该项目采用TypeScript编写，其核心技术特点在于使用真实浏览器环境而非无头浏览器或爬虫库，从而避免了反爬机制，并且无需重新登录即可操作已登录账号。适用于需要以个人身份自动化执行网页操作的场景，例如社交媒体管理、在线学习辅助以及数据分析等。",2,"2026-06-11 03:50:50","high_star"]