agent-browser
View on GitHubBrowser automation CLI for AI agents
Rust-based CLI that gives AI agents headless Chrome control: open pages, snapshot the accessibility tree with element refs, click/fill/type, read agent-friendly text, screenshot, and invoke WebMCP page tools. Ships an MCP profile and skills, installable via npm, Homebrew, or Cargo.
Use Cases
Let AI agents navigate and interact with websites via CLICapture annotated screenshots and accessibility trees for agentsFill forms and click elements by stable refs from snapshotsExtract agent-readable page text and documentationInvoke WebMCP page-provided tools from an agentConnect to existing Chrome over CDP for debuggingRun browser automation without Playwright or PuppeteerGenerate PDFs and full-page screenshots in agent workflows
Built With
- Language
- Rust
- Frameworks
- Node.js · Rust · Chrome DevTools Protocol · pnpm · Docker
Tags
browser-automation · cli · rust · headless-chrome · cdp · accessibility-tree · screenshots · webmcp · mcp · agent-tooling · automation · web-scraping · pdf-export · skills · evals