browser-use-skill
Claude Code skill for AI-powered browser automation with two modes
browser-use-skill is a Claude Code skill that wraps the official browser-use Python library to enable AI-powered browser automation. It offers two modes: Direct Mode, where Claude controls the browser step-by-step using Vision without requiring an external LLM API key, and Subagent Mode, where complex tasks are delegated to autonomous Claude Code subagents. The skill maintains persistent browser sessions across multiple tool calls via a local server.
A local Python server maintains a persistent browser session, exposing tools for navigation, clicking, typing, and screenshots that Claude can call sequentially without losing state.
developers building AI-powered browser automation workflows with Claude Code
Background.
- Status
- launched
- Business model
- open-source
Founders
- dalbit-mirgithub
Similar projects.
Editorial take on the space this project sits in — momentum signals, adjacent moves, our call on whether the wedge is real. Get pinged when we publish a new read or when the landscape shifts.
Have a take on this space?
Tell us what you’d build differently, where you think the incumbents miss, or what we’ve gotten wrong about this project. Comments + reactions are coming soon.