Library
88,213 resources from 11 public sources. 54 checked by hand. The rest are indexed automatically and not reviewed.
Results
Looks like a goal we have a starting path for: I want to build an AI agent
-
mearl
SkillsA browser work environment for AI agents: automate, debug, and verify in the browser of your choice.
Source for mearl (opens github.com) git clone https://github.com/mearlhq/mearl.git -
ruin-everything
MCP serversA local Chrome runtime for browser agents. Reuse logins, run tasks in the background, handle supported verification steps, and keep submission receipts.
Source for ruin-everything (opens github.com) git clone https://github.com/yuki4266/ruin-everything.git -
agnt
MCP serversMCP Server for Development Tooling - Project Detection, Process Management, and Reverse Proxy with Frontend Instrumentation
Source for agnt (opens github.com) git clone https://github.com/standardbeagle/agnt.git -
imprint
MCP serversRecord a browser workflow once, get a deterministic MCP tool your AI agent can call. Zero tokens at runtime. Postman for AI agents.
Source for imprint (opens github.com) git clone https://github.com/ashaychangwani/imprint.git -
Browsentic
MCP serversAgentic browser harness: your agent CLI in your real, signed-in browser, from a side panel or MCP.
Source for Browsentic (opens github.com) browsentic@0.9.2 -
nuphus-mcp
MCP serversDesktop automation MCP server — computer use for any AI agent: control screen, windows, mouse/keyboard, and Chrome via Model Context Protocol (stdio). Not a DSH plugin — DSH users install dsh-nuphus-mcp.
Source for nuphus-mcp (opens github.com) git clone https://github.com/mrpulor-gh/nuphus-mcp.git -
agent-workspace-linux
MCP serversIsolated Linux desktop workspaces for AI agents — a hidden, agent-owned desktop and browser over MCP, so an agent can do GUI and web work without touching your real desktop.
Source for agent-workspace-linux (opens github.com) git clone https://github.com/agent-sh/agent-workspace-linux.git -
open-browser-use
MCP servers🔮 Platform-neutral Browser Use for AI agents: real Chrome automation with a CLI + SDKs, no lock-in, dead simple.
Source for open-browser-use (opens github.com) git clone https://github.com/iFurySt/open-browser-use.git -
liliumai
Agents & frameworksLilium AI: The ultimate personal AI agent framework for autonomous computer control. Featuring browser automation, shell execution, and multi-channel integration (WeChat/Telegram/Discord).
Source for liliumai (opens github.com) git clone https://github.com/beidald/liliumai.git -
axuse
MCP serversAccessibility-first browser & computer use for AI agents — one MCP server for every harness (Codex, Claude Code, OpenCode, Antigravity). Instruments the DOM/accessibility tree via CDP instead of screenshot loops: diffed snapshots, stable element refs, trusted input, typed errors. Zero dependencies
Source for axuse (opens github.com) git clone https://github.com/ctoldigital/axuse.git -
vibesurfer
SkillsA browser for LLMs, not humans.
Source for vibesurfer (opens github.com) git clone https://github.com/frane/vibesurfer.git -
pagesift
AI repositoriesLocal-first web & document ingestion engine in Rust — HTTP/browser capture, layout-aware PDF/Office/OCR parsing, persistent sync, verifiable data graphs, and evidence-grounded AI context.
Source for pagesift (opens github.com) git clone https://github.com/YOLOW92/pagesift.git -
octoweb
MCP serversKeyboard-first AI browser for macOS — WebKit + Rust, built-in AI sidebar, isolated workspaces, and an MCP server that lets your agents drive the web.
Source for octoweb (opens github.com) git clone https://github.com/Muvon/octoweb.git -
BossConsole
MCP serversOpen-source, multi-platform harness for AI agents - a native, multi-threaded operator's console (JVM, not Electron) to run Claude Code, Codex, Gemini or OpenCode with a real browser, terminal, editor, secrets & 200+ MCP tools. Built for enterprises, science & research.
Source for BossConsole (opens github.com) git clone https://github.com/risa-labs-inc/BossConsole.git -
browser-bridge
MCP serversDrive your real, logged-in Chrome from an AI agent (Claude Code / Codex) over MCP - 55 tools for browsing, DevTools-grade network capture, and web-security testing.
Source for browser-bridge (opens github.com) git clone https://github.com/vitalysim/browser-bridge.git -
agent-qa
MCP serversOpen-source self-improving QA agent for software teams. A test harness with memory. Write tests in natural language for web and mobile. agent-qa learns from every run, adapts to UI changes, and catches regressions before you ship.
Source for agent-qa (opens github.com) git clone https://github.com/vostride/agent-qa.git -
sctl
MCP serversControl your real browser and ScriptCat from the CLI, AI clients (MCP) and scripts: tabs, browser data, page automation, debugging, raw CDP, and userscript management — one cross-platform binary.
Source for sctl (opens github.com) git clone https://github.com/scriptscat/sctl.git -
plasmate
MCP serversBrowser engine for agents. HTML in, structured Semantic Object Model out. Native MCP and AWP with a supported CDP subset. Apache-2.0.
Source for plasmate (opens github.com) git clone https://github.com/plasmate-labs/plasmate.git -
OpenAdapt
Agents & frameworksFor EHRs, payer portals, and any app your team clicks through. A person or an agent shows the task; OpenAdapt turns it into a local program that repeats the steps with no model calls, checks each save, and stops when something doesn't match.
Source for OpenAdapt (opens github.com) git clone https://github.com/OpenAdaptAI/OpenAdapt.git -
agentes-chat
Agents & frameworksOpen-source, self-hosted Slack alternative where your team and AI agents work together in channels and DMs. Agents run on Claude, ChatGPT/Codex, Gemini or Grok, with MCP tools, their own computer and browser, routines, memory and approvals. Web, desktop and phone apps. Docker Compose + Postgres. AGPL-3.0.
Source for agentes-chat (opens github.com) git clone https://github.com/agentes-chat/agentes-chat.git -
codex-browser-bridge
MCP serversMCP server and CLI for reusing a local Chromium browser session (Browser Use) and Codex cua_repl (Computer Use).
Source for codex-browser-bridge (opens github.com) git clone https://github.com/DeliciousBuding/codex-browser-bridge.git -
yeli
SkillsA browser-based Markdown editor for people and AI agents. Portable HTML documents with offline editing. Online editor in development. | 页里:人和 AI 都好用的 Markdown 编辑器。离线单文件,在线编辑器开发中。
Source for yeli (opens github.com) git clone https://github.com/whitesmell/yeli.git -
browser-copilot
MCP serversAI side-panel assistant for Chrome. Reads the active page and acts on it with your approval — click, type, switch tabs, fill forms. Multi-LLM, reusable Skills, recordable workflows, MCP bridge to Claude Code / Codex / Trae, optional Node/Playwright runner.
Source for browser-copilot (opens github.com) git clone https://github.com/dcc123456/browser-copilot.git -
e2e-skills
SkillsPlaywright and Cypress E2E testing skills for AI agents. Generate Playwright tests, review test quality, and debug failures in Claude Code and Codex.
Source for e2e-skills (opens github.com) git clone https://github.com/voidmatcha/e2e-skills.git
Indexed entries aren't reviewed or tested by InnerVault. Details come straight from each source. Anything that can reach your accounts, files or browser can do damage, so read the source before you install it.