Scout: the open-source browser agent, by Outcrawl. Mirrored automatically from Outcrawl's own repository.
Open-source, self-hosted alternative to OpenAI Dots, Grok Bot. Built to be undetectable by anti-bot systems.
JavaScript in-page GUI agent. Control web interfaces with natural language.
The SDK to extract data and interact with any site on the web. Get started with Claude Code, Codex, Eve, Mastra, and more.
Turn any website into a structured API. Extract, automate, search and monitor the web.
Open-source AI agent for web automation. Use your real, logged-in browser to automate daily tasks.
๐ The open-source Agentic browser; alternative to ChatGPT Atlas, Perplexity Comet, Dia.
๐ฅ Open Source Browser API for AI Agents & Apps. Steel Browser is a batteries-included browser sandbox that lets you automate the web without worrying about infrastructure.
Open-source AI coworkers that each get a computer of their own: a browser, files and tools, with every action decided before it happens and recorded after. Bring any AG-UI agent.
Atomic Agent is a local-first AI agent. Runs open-weight models on your own machine via llama.cpp.
Self-learning agent browser
The undetected self-hosted browser automation platform. Powered by Camoufox (Firefox) for 0% detection rates. Built for speed, privacy, and scalability.
Browser and desktop automation with an MCP server for AI assistants. Record macros or write JavaScript to automate websites, extract data, and control desktop apps with OCR and image recognition.
A search-first, profile-aware reverse-engineering workspace for AI agents โ 735 tools across 36 domains in one MCP server.
A JavaScript library for generating random user agents with data that's updated daily.
Browser automation clicks buttons. OpenTabs calls APIs.
Chrome sidebar extension that lets DeepSeek Harness operate your browser directly, no vision capabilities required. ไธๆฌพ Chrome ไพง่พนๆ ๆฉๅฑ็จๅบ๏ผๅฏ่ฎฉ DeepSeek Harness ็ดๆฅๆๆงๆจ็ๆต่งๅจ๏ผๆ ้่ง่ง่ฝๅใ
Stack blocks visually to build complex browser workflows and execute them via API
Local-first open-source multi-profile fingerprint browser based on Chromium and Electron
The CLI for AI agents to control Chrome. Zero config, agent-agnostic, battle-tested.
Browser automation and computer use for AI agents. One CLI and MCP server for your signed-in browser, native macOS apps, and real iPhones.
Grasp helps you manage multiple projects and parallel coding agents in one visual workflow, so humans can understand faster and ship more.
Render a real Chromium view inside a Herdr pane and drive it over CDP.
Agentic browser control driven by your own Claude / Cursor subscriptions
Your browser. Ready to act. An MIT-licensed desktop browser agent for website tasks, cited research, and supported games.
Tabbit Browser plugins for Deepseek Harness
Talk, and a team of bots does the work in a real browser, a shell and your files while she keeps talking. Open source, runs on your machine, on your ChatGPT plan or an OpenAI key. npx thursday-agent
An open-source travel agent that works on real websites: plan trips, book in a browser beside the chat, stop at the payment page. Desktop app for macOS, Windows and Linux.
Jev Browser โ indexed browser automation. Jev chooses the control, Playwright acts. A CodexQA skill.
Browser extension for the DeepSeek Harness desktop app. The model reads the page you are on as text with numbered controls and acts on it โ click, type, scroll, navigate, manage tabs โ and, when you switch image recognition on, looks at an image you point at. It asks before acting, keeps passwords in the page, and talks to your own desktop; the one
Open-source autonomous AI browser agent Chrome extension running locally in your authenticated browser. Researches web data, fills forms, automates workflows, and streams live results using active logins.
An agentic browser harness: an AI side panel & orb in your own logged-in Chrome, Edge, Arc, Brave or even Firefox run by the agent/subscription you already use (Claude Code, Codex, Antigravity). Free, open source, no API key. Also an MCP server.
Rebuild a local twin of your app on every push, replay approved business journeys, and report a GitHub check.
The visual layer of your agent harness. Renders React components and Three.js scenes from source, with structure and evidence behind every frame โ so agents can see, verify, and fix what they built.
Jev for Chrome: drives the tab you are looking at with TypeSafe Jev, a sub-second decision model. Community port of browser-use/jev-ultrafast, not affiliated with TypeSafe.
Pay-per-call APIs and MCP services for AI agents: document OCR, document analysis, media transcription, browser sessions, and agent inboxes. USDC on Base over x402, no API keys, receipt-backed results.
BrowserClaw - Control your everyday Chrome browser from AI agents, without losing logins or focus
Browser-authenticated Fetch and existing-browser Connect for AI agents โ use Chrome or Edge login state with site adapters, agent-browser, Browser Use, and Playwright CLI.
The desktop home for your coding agent โ Claude Code or Codex. A persistent chat, a real browser it drives on your own logins, your phone in the loop, and routines on a schedule. Local, on your own subscription, open source. macOS.
AI side-panel assistant for Chrome. Reads the active page and acts on it with your approval โ click, type, switch tabs, fill forms. Multi-LLM, reusable Skills, recordable workflows, MCP bridge to Claude Code / Codex / Trae, optional Node/Playwright runner.
Human escalation for Solari browser agents โ hand off the live session, solve the blocker, resume in the same session.
The desktop AI agent for the real web. Built-in fingerprint browsers, saved logins, multi-agent orchestration, and reusable Browser Skills beyond local coding.
Build and reuse WebMCP tools in a macOS desktop workspace for DeepSeek Harness.
Fast browser and macOS computer use for MCP agents. Local execution, TypeSafe Jev decisions, verified outcomes.