The top browser and computer-use agents, numbered by AgentPick Score, each with its source. Ranked by AgentPick Score, our 0–100 data score. Last updated 15 Oct 2026.
Open-source library that lets AI agents control a web browser. This release adds Browser Use toolsets for Claude. The notes say it…
★ 118k · released 7 Oct 2026 · 365 contributors · MIT
Scale computer-use 2.0 with open-source drivers, cross-OS fleets, and benchmarks for training, evaluation, and data generation.
★ 30k · released 14 Oct 2026 · 181 contributors · MIT
The SDK to extract data and interact with any site on the web. Get started with Claude Code, Codex, Eve, Mastra, and more.
★ 26k · released 14 Oct 2026 · 88 contributors · MIT
Microsoft's Playwright MCP server lets agents drive a real browser. 0.0.83 lets browser_find save matches to a file, and navigate …
★ 38k · released 28 Sept 2026 · 68 contributors · Apache-2.0
Stealth Chromium that passes every bot detection test. Drop-in Playwright replacement with source-level fingerprint patches. 50+ t…
★ 32k · released 14 Oct 2026 · 23 contributors · MIT
Automate browser based workflows with AI
★ 23k · released 14 Oct 2026 · 110 contributors · AGPL-3.0
See interfaces like a human. E2E testing in natural language.
★ 15k · released 14 Oct 2026 · 95 contributors · MIT
Make Any Website into CLI & Use your logged-in browser by AI agent.
★ 30k · released 30 Aug 2026 · 273 contributors · Apache-2.0
Best headless browser for AI agents. Lite, Fast, High-Compatibility. Built in Rust
★ 18k · released 12 Oct 2026 · 15 contributors · Apache-2.0
The fastest browser for AI agents to run browser automation, built for sharing your logged-in browser state with your AI agents, l…
★ 17k · released 12 Oct 2026 · 11 contributors · MIT
The Open-Source Multimodal AI Agent Stack: Connecting Cutting-Edge AI Models and Agent Infra
★ 39k · released 8 Oct 2026 · 50 contributors · Apache-2.0
🌐 The open-source Agentic browser; alternative to ChatGPT Atlas, Perplexity Comet, Dia.
★ 14k · released 13 Oct 2026 · 24 contributors · AGPL-3.0
Automate your mobile devices with natural language commands - an LLM agnostic mobile Agent 🤖
★ 9.6k · released 14 Oct 2026 · 32 contributors · MIT
JavaScript in-page GUI agent. Control web interfaces with natural language.
★ 29k · released 6 Sept 2026 · 36 contributors · MIT
Open-source AI coworkers that each get a computer of their own: a browser, files and tools, with every action decided before it ha…
★ 6.5k · released 12 Oct 2026 · 42 contributors · MIT
Browser Harness | Self-healing harness that enables LLMs to complete any task.
★ 18k · released 4 Sept 2026 · 89 contributors · MIT
Stealth headless browser for AI agents — bypass Cloudflare, bot detection, and anti-scraping. Drop-in Puppeteer/Playwright replace…
★ 12k · released 5 Oct 2026 · 57 contributors · MIT
Open-Source Chrome extension for AI-powered web automation. Run multi-agent workflows using your own LLM API key. Alternative to O…
★ 14k · released 2 Oct 2026 · 26 contributors · Apache-2.0
Chrome extension & CLI to let agents control your browser. Runs Playwright snippets in a stateful sandbox. Available as CLI or MCP
★ 4k · released 7 Oct 2026 · 7 contributors · MIT
Browser4 — an AI-native browser engine for autonomous agents, intelligent extraction, and large-scale web automation.
★ 1.2k · released 14 Oct 2026 · 9 contributors · Apache-2.0Browser Use 0.13.11Verdict: Worth watching for Claude users, but per its own release notes the new integration can't be used with today's public Anthropic SDK.
Test notes: The verdict is based on the linked public release notes only.
Pricing: Open source
Playwright MCP 0.0.83Verdict: A reliable, scriptable way to give any MCP-capable agent a browser. A good pairing with coding agents for end-to-end checks.
Test notes: The verdict is based on the linked public sources only.
Pricing: Open source
Our top pick: Browser Use, because it leads browser agents on AgentPick Score (92/100) with 118k GitHub stars, a release on 7 Oct 2026 and 365 contributors. Runners-up: Cua (87) and Stagehand (86).
Every tool gets an AgentPick Score from 0 to 100, computed from public data: GitHub stars, how recently it was committed to and released, release cadence, contributors, license openness and docs presence. Tools we've also run ourselves carry a Hands-on tested badge. Rankings can't be bought.
The week's new agent tools with a verdict on each, every week. Free. We'll email you a link to confirm, and you can unsubscribe any time. See our Privacy Policy.