{"name":"Loadbay","description":"Catalog of harnesses for AI agents, across domains.","repo":"https://github.com/alessandrorodi/loadbay","count":29,"domains":{"trading":"Trading","coding":"Coding & DevOps","browser":"Browser & computer","productivity":"Productivity","data":"Data & search","social":"Social & comms","health":"Health","science":"Research & science","gaming":"Gaming","media":"Voice & media","robotics":"Robotics"},"traits":{"mcp":"Exposes an MCP server","keys":"Needs API keys / auth","sandbox":"Runs sandboxed","guard":"Guardrails / limits"},"harnesses":[{"slug":"nanobrowser","name":"nanobrowser","author":"nanobrowser","domain":"browser","integrations":["Chrome"],"language":"TypeScript","license":"Apache-2.0","stars":13307,"traits":["oss","keys"],"summary":"An open-source Chrome extension that runs multi-agent web-automation workflows right in your browser.","repo":"https://github.com/nanobrowser/nanobrowser","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/nanobrowser","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"browser-use","name":"browser-use","author":"browser-use","domain":"browser","integrations":["Chrome","Any website"],"language":"Python","license":"MIT","stars":99265,"traits":["oss","keys"],"summary":"Make any website usable by an agent. It drives a real browser to click, type, and finish tasks online — the most-starred browser harness by a wide margin.","repo":"https://github.com/browser-use/browser-use","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/browser-use","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"agent-s","name":"Agent-S","author":"simular-ai","domain":"browser","integrations":["Computer"],"language":"Python","license":"Apache-2.0","stars":11865,"traits":["oss"],"summary":"An open framework that lets an agent use a computer the way a person does: read the screen, move the mouse, type.","repo":"https://github.com/simular-ai/Agent-S","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/agent-s","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"cua","name":"cua","author":"trycua","domain":"browser","integrations":["Computer","macOS"],"language":"Python","license":"MIT","stars":18396,"traits":["oss","sandbox"],"summary":"Infrastructure for computer-use agents: sandboxes, SDKs, and benchmarks so an agent can drive a whole desktop without escaping it.","repo":"https://github.com/trycua/cua","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/cua","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"bytebot","name":"Bytebot","author":"bytebot-ai","domain":"browser","integrations":["Computer"],"language":"TypeScript","license":"Apache-2.0","stars":11000,"traits":["oss","sandbox","keys"],"summary":"A self-hosted AI desktop agent that automates computer tasks inside its own containerized desktop.","repo":"https://github.com/bytebot-ai/bytebot","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/bytebot","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"self-operating-computer","name":"self-operating-computer","author":"OthersideAI","domain":"browser","integrations":["Computer"],"language":"Python","license":"MIT","stars":10200,"traits":["oss","keys"],"summary":"A framework that lets a multimodal model operate a computer by looking at the screen and moving the mouse.","repo":"https://github.com/OthersideAI/self-operating-computer","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/self-operating-computer","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"skyvern","name":"skyvern","author":"Skyvern-AI","domain":"browser","integrations":["Chrome","Playwright"],"language":"Python","license":"AGPL-3.0","stars":21923,"traits":["oss","keys","guard"],"summary":"Automates browser-based workflows using LLMs and computer vision to operate websites without site-specific scripts.","repo":"https://github.com/Skyvern-AI/skyvern","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/skyvern","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"stagehand","name":"stagehand","author":"browserbase","domain":"browser","integrations":["Chrome","Playwright"],"language":"TypeScript","license":"MIT","stars":23147,"traits":["oss","keys","sandbox"],"summary":"SDK for browser agents that adds act, extract, and observe primitives on top of Playwright for AI-driven web automation.","repo":"https://github.com/browserbase/stagehand","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/stagehand","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"lavague","name":"LaVague","author":"lavague-ai","domain":"browser","integrations":["Selenium","Playwright"],"language":"Python","license":"Apache-2.0","stars":6370,"traits":["oss","keys"],"summary":"Large Action Model framework that turns natural-language objectives into executable web automation for AI agents.","repo":"https://github.com/lavague-ai/LaVague","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/lavague","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"steel-browser","name":"steel-browser","author":"steel-dev","domain":"browser","integrations":["Chrome","Puppeteer"],"language":"TypeScript","license":"Apache-2.0","stars":7172,"traits":["oss","keys","sandbox"],"summary":"Open-source browser API and sandbox that lets AI agents automate the web without managing browser infrastructure.","repo":"https://github.com/steel-dev/steel-browser","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/steel-browser","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"notte","name":"notte","author":"nottelabs","domain":"browser","integrations":["Chrome","Playwright"],"language":"Python","license":"Other","stars":1975,"traits":["oss","keys","sandbox"],"summary":"Framework to build web agents and deploy serverless web automation functions on managed browser infrastructure.","repo":"https://github.com/nottelabs/notte","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/notte","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"open-interpreter","name":"open-interpreter","author":"openinterpreter","domain":"browser","integrations":["Shell","OS"],"language":"Python","license":"Apache-2.0","stars":64030,"traits":["oss","keys"],"summary":"Lets LLMs run code and control the local computer through a natural-language interface for OS-level automation.","repo":"https://github.com/openinterpreter/open-interpreter","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/open-interpreter","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"openadapt","name":"OpenAdapt","author":"OpenAdaptAI","domain":"browser","integrations":["Desktop","OS"],"language":"Python","license":"MIT","stars":1616,"traits":["oss","keys"],"summary":"Generative RPA that records desktop screen and input activity and replays it with multimodal models to automate GUI tasks.","repo":"https://github.com/OpenAdaptAI/OpenAdapt","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/openadapt","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"midscene","name":"midscene","author":"web-infra-dev","domain":"browser","integrations":["Chrome","Playwright"],"language":"TypeScript","license":"MIT","stars":13761,"traits":["oss","keys"],"summary":"Vision-driven UI automation that drives web and mobile interfaces from natural language for AI agents.","repo":"https://github.com/web-infra-dev/midscene","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/midscene","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"browser-use-web-ui","name":"web-ui","author":"browser-use","domain":"browser","integrations":["Chrome","Playwright"],"language":"Python","license":"MIT","stars":16098,"traits":["oss","keys"],"summary":"Browser-based UI for running web-automation agents with support for custom models and persistent browser sessions.","repo":"https://github.com/browser-use/web-ui","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/browser-use-web-ui","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"omniparser","name":"OmniParser","author":"microsoft","domain":"browser","integrations":["Desktop","OS"],"language":"Python","license":"Other","stars":24915,"traits":["oss"],"summary":"Screen-parsing tool that converts UI screenshots into structured elements to enable pure vision-based GUI agents.","repo":"https://github.com/microsoft/OmniParser","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/omniparser","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"ufo","name":"UFO","author":"microsoft","domain":"browser","integrations":["Windows","OS"],"language":"Python","license":"MIT","stars":9022,"traits":["oss","keys"],"summary":"UI-focused agent that operates Windows applications via natural language using GUI and API actions.","repo":"https://github.com/microsoft/UFO","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/ufo","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"open-interface","name":"Open-Interface","author":"AmberSahdev","domain":"browser","integrations":["Desktop","OS"],"language":"Python","license":"GPL-3.0","stars":2688,"traits":["oss","keys"],"summary":"Controls any computer using LLMs by simulating keyboard and mouse to complete user-specified tasks across apps.","repo":"https://github.com/AmberSahdev/Open-Interface","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/open-interface","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"maxun","name":"maxun","author":"getmaxun","domain":"browser","integrations":["Chrome","Playwright"],"language":"TypeScript","license":"AGPL-3.0","stars":15924,"traits":["oss","keys"],"summary":"No-code platform that turns websites into structured APIs through browser-based scraping and AI data extraction.","repo":"https://github.com/getmaxun/maxun","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/maxun","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"playwright-mcp","name":"playwright-mcp","author":"microsoft","domain":"browser","integrations":["Chrome","Playwright"],"language":"TypeScript","license":"Apache-2.0","stars":34000,"traits":["mcp","oss"],"summary":"Microsoft's Playwright MCP server — drive a real browser (navigate, click, fill, assert) from any agent.","repo":"https://github.com/microsoft/playwright-mcp","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/playwright-mcp","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"anthropic-quickstarts","name":"Anthropic computer-use demo","author":"anthropics","domain":"browser","integrations":["Claude API","Docker","Bedrock","Vertex"],"language":"Python","license":"MIT","stars":17000,"traits":["oss","keys","sandbox"],"summary":"Anthropic official computer-use reference: a containerized Linux desktop where Claude controls the GUI via screenshots and tool calls.","repo":"https://github.com/anthropics/anthropic-quickstarts","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/anthropic-quickstarts","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"chrome-devtools-mcp","name":"Chrome DevTools MCP","author":"ChromeDevTools","domain":"browser","integrations":["MCP","Chrome","Puppeteer"],"language":"TypeScript","license":"Apache-2.0","stars":42000,"traits":["oss","mcp"],"summary":"Official Chrome DevTools MCP server letting an agent control and inspect a live Chrome browser for automation and debugging.","repo":"https://github.com/ChromeDevTools/chrome-devtools-mcp","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/chrome-devtools-mcp","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"ui-tars-desktop","name":"UI-TARS Desktop","author":"bytedance","domain":"browser","integrations":["MCP","UI-TARS","HuggingFace"],"language":"TypeScript","license":"Apache-2.0","stars":36700,"traits":["oss","mcp","keys"],"summary":"ByteDance multimodal agent stack (Agent TARS + UI-TARS Desktop) that controls computer and browser operators via natural language.","repo":"https://github.com/bytedance/UI-TARS-desktop","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/ui-tars-desktop","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"browseros","name":"BrowserOS","author":"browseros-ai","domain":"browser","integrations":["Chromium","local LLMs","MCP"],"language":"TypeScript","license":"AGPL-3.0","stars":11400,"traits":["oss","mcp","keys"],"summary":"Open-source agentic web browser (Chromium fork) that runs AI agents natively inside the browser.","repo":"https://github.com/browseros-ai/BrowserOS","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/browseros","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"webarena","name":"WebArena","author":"web-arena-x","domain":"browser","integrations":["Docker","Playwright","OpenAI"],"language":"Python","license":"Apache-2.0","stars":1500,"traits":["oss","keys","sandbox"],"summary":"Self-hostable realistic web environment (e-commerce, forums, GitLab, CMS) for building and benchmarking autonomous web agents.","repo":"https://github.com/web-arena-x/webarena","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/webarena","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"webvoyager","name":"WebVoyager","author":"MinorJerry","domain":"browser","integrations":["Selenium","GPT-4V","OpenAI"],"language":"Python","license":"Other","stars":1100,"traits":["oss","keys"],"summary":"End-to-end multimodal web agent that completes instructions on real live websites with set-of-mark prompting.","repo":"https://github.com/MinorJerry/WebVoyager","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/webvoyager","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"opencua","name":"OpenCUA","author":"xlang-ai","domain":"browser","integrations":["vLLM","HuggingFace","OSWorld"],"language":"Python","license":"Apache-2.0","stars":784,"traits":["oss","keys"],"summary":"Open foundation models and framework for computer-use agents, including dataset, benchmark, and end-to-end CUA models.","repo":"https://github.com/xlang-ai/OpenCUA","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/opencua","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"agent-e","name":"Agent-E","author":"EmergenceAI","domain":"browser","integrations":["AutoGen","Playwright","OpenAI"],"language":"Python","license":"MIT","stars":1200,"traits":["oss","keys"],"summary":"Agent-driven browser automation built on the AutoGen framework for autonomous web task execution via natural language.","repo":"https://github.com/EmergenceAI/Agent-E","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/agent-e","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}},{"slug":"playwright-mcp-ea","name":"Playwright MCP (ExecuteAutomation)","author":"executeautomation","domain":"browser","integrations":["MCP","Playwright","Cursor","Cline"],"language":"TypeScript","license":"MIT","stars":5600,"traits":["oss","mcp"],"summary":"Popular community Playwright MCP server enabling agents to automate browsers and APIs, with screenshots and codegen.","repo":"https://github.com/executeautomation/mcp-playwright","community":false,"tip":{"method":"x402","network":"eip155:8453","endpoint":"/api/tip/playwright-mcp-ea","note":"GET with an x402 client to tip the author; returns 402 with payment requirements until paid."}}]}