openbrowser and nanobrowser

The tools are competitors, both offering open-source browser automation solutions for AI agents to interact with web applications, but with different underlying architectures—one as an autonomous toolkit and the other as a Chrome extension.

openbrowser
59
Established
nanobrowser
52
Established
Maintenance 13/25
Adoption 10/25
Maturity 16/25
Community 20/25
Maintenance 6/25
Adoption 10/25
Maturity 16/25
Community 20/25
Stars: 9,228
Forks: 837
Downloads:
Commits (30d): 2
Language: TypeScript
License: MIT
Stars: 12,440
Forks: 1,296
Downloads:
Commits (30d): 0
Language: TypeScript
License: Apache-2.0
No Package No Dependents
No Package No Dependents

About openbrowser

ntegrals/openbrowser

Let AI agents browse the web. An autonomous toolkit for browser-based AI agents.

Built on Playwright with pluggable LLM integration (OpenAI, Anthropic, Google via Vercel AI SDK), agents autonomously navigate websites by analyzing DOM snapshots, deciding actions, and executing commands like click, type, and extract. Includes production features like stall detection, cost tracking, sandboxed execution with resource limits, session replay recording, and an interactive REPL for debugging — all available as a TypeScript library or CLI.

About nanobrowser

nanobrowser/nanobrowser

Open-Source Chrome extension for AI-powered web automation. Run multi-agent workflows using your own LLM API key. Alternative to OpenAI Operator.

Built on a **multi-agent architecture** with specialized Navigator and Planner agents that communicate to handle complex web tasks, the extension supports flexible LLM provider integration (OpenAI, Anthropic, Gemini, Ollama, and OpenAI-compatible APIs) allowing per-agent model assignment for cost optimization. Executes entirely within your browser using Chrome's extension APIs with real-time conversation history and follow-up question capabilities, eliminating cloud dependencies while maintaining full local privacy.

Scores updated daily from GitHub, PyPI, and npm data. How scores work