whereismyai
browser-control
browser-control is a tiny, fast Rust CLI that drives a real browser over the Chrome DevTools Protocol for coding agents.
Browser-control exposes a browser as a small, debuggable pipe with page snapshots, stable element refs, raw CDP, a background event watcher, and shell-runnable scripts. It is built for coding agents that already have their own planning, retry, and file-editing loops, giving them hands and eyes without baking brittle intelligence into the tool. Every capability is a subcommand printing compact text or JSON, allowing any agent that can run a shell command to drive a browser. It supports local Chrome or remote sessions on cloud providers like Browser Use, Steel, Hyperbrowser, and Browserbase using the same commands. A hidden daemon keeps in-memory event, network, and console rings, automatically writing compact traces on command failure.
browser-control at a glance
- Best for: Software engineer
- Runs on: CLI, self-hosted, open source
- Pricing: Free
- Released: December 2016
- Category: AI agent tool
- Website: github.com/keon/browser-control
Key features
- Drives local or remote browsers over Chrome DevTools Protocol
- Provides stable element handles via snapshot for LLM workflows
- Exposes raw CDP escape hatch for advanced use cases
- Keeps in-memory event, network, and console rings
- Writes compact error traces automatically on failure
- Supports shell-native subcommands printing text or JSON
- Integrates with cloud browser providers via environment variables
Use it for
- Driving a browser from a coding agent via shell commands
- Capturing page snapshots with stable element references
- Executing raw CDP commands and custom JavaScript evaluations
- Running remote browser sessions on cloud browser providers
Companions
Tools that articles about browser-control recommend using with it, each with its job.
- Playwright — Hand-coded automation for the fixed, high-volume parts.
- Ollama — Any OpenAI-compatible local model, at no cost per task.
- LangChain — Compose exactly the agent your use case needs from model, tools, prompt, and middleware.
- CrewAI — Gives CrewAI agents a real browser.
Alternatives to browser-control
- Dev Browser — Persistent browser tabs keep state across scripts so agents do not restart.
- Browser Use — Managed remote browsers built for protected sites at $0.02 an hour.
- BrowserOS — Runs entirely locally and lets AI agents use real user logins to complete tasks.
- Webwright — Enables code-based web automation with reusable and rerunnable scripts.
- jevgrep — Reduces coding agent costs by providing targeted, relevant repository context.
- Steel — Run scalable cloud browser sessions with a simple developer API.
- Pw Repl — Shares a single browser session between users and AI agents in real time.
- Agent Chrome Relay — Runs in background tabs using real logins without stealing browser focus.
- Webcmd — Learns websites as agents use them to compile faster and cheaper commands.
- AG-UI — Standardizes agent-user interactions and generative UI for applications.
- OpenAgentsControl — Generates code matching your patterns and requires approval before execution.
- Browser Mini — It combines a native Chromium browser with a built-in REST server for automation.
All browser-control alternatives