{"repo":"idan-rubin/browserclaw-agent","free":true,"listed":false,"github":"https://github.com/idan-rubin/browserclaw-agent","clone":"git clone https://github.com/idan-rubin/browserclaw-agent.git","description":"AI browser automation conductor — type a prompt, watch it happen live, get a reusable skill","language":"TypeScript","stars":11,"topics":["ai","ai-agent","browser-automation","chrome","llm","nextjs","playwright","typescript","web-scraping","browserclaw"],"license":"MIT","category":"scrapers-browser-automation","readme_excerpt":"browserclaw-agent The AI agent for browserclaw. Layered, not bundled Three separate layers, not one monolith: - ⚡ LLM — the electricity. Your choice: Claude, GPT, Gemini, local. No lock-in. - 😎 The agent — the driver. Reasoning, obstacle recovery, learned skills. That's this project. - 🏎️ BrowserClaw — the vehicle. Snapshots, element refs, browser control. Standalone npm library. browser-use welds these into one package. We keep them as separate, swappable layers — drop the library into your own agent, pair this agent with any LLM, or run the whole stack as-is. vs browser-use Different lineage, different design. OpenClaw took the Playwright MCP approach — Microsoft's snapshot-and-ref pattern — implemented it locally, and refined it into browserclaw, a standalone npm library. This agent rides on that library. browser-use rolled its own Python stack as one bundled package. browserclaw browser-use :-------------------------- :----------------: :----------------: Standalone browser library :white check mark: :x: Works with any agent :white check mark: :heavy minus sign: Auto-learns skills per site :white check mark: :x: Solves anti-bot in OSS :white check mark: :heavy minus sign: TypeScript / Node :white check mark: :x: :white check mark: = Yes  :heavy minus sign: = Partial  :x: = No What the agent does The agent reads an accessibility snapshot of the page, decides what to do next, and executes the action. Up to 100 steps per run. It maintains a memory scratchpad acro","default_branch":null,"files":null,"tree":[],"storefront":"/r/idan-rubin","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/idan-rubin/browserclaw-agent/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}