浏览器自动化

Model Context Protocol Servers

Chrome DevTools for coding agents

Playwright MCP server

🌐 Make websites accessible for AI agents. Automate tasks online with ease.

Allow LLMs to control a browser with Browserbase and Stagehand

Open-source MCP server for secure, low-latency cloud-browser automation on Kernel.

Drive your real, logged-in Chrome from any AI agent (Claude Code, Cursor, VS Code) — works where headless dies. Reads emailed login codes from your Gmail, solves CAPTCHAs, 34 tools. MIT, local-only.

Native Swift macOS automation — 143 tools: AX, EventKit, voice, browser DOM, undo.

Human-to-AI code review bridge. Review UI prototypes in the browser, AI agents fix the code via MCP.

Code-aware browser testing agent — 13 MCP tools for AI editors (Claude Code, Cursor, Windsurf). Reads your codebase, opens Playwright, tests everything, reports with screenshots.

The persistent home a stateless AI agent can't host for itself — register once, resume your whole self in one call; real browser; cited deep research; hire real humans for physical-world tasks; free keyless chat + ID portraits. MCP + REST, x402 on Base, free to start.

SiFR (Salience-Indexed, Flat Relations) format specification for Element to LLM (E2LLM): the structured page representation your AI reads - format spec, taxonomy, real captures, MCP server manifest, client configs, and the model skill.

Ultra-lightweight headless browser for AI agents. Single Rust binary, 4 language SDKs (Python/JS/C#/Java).

MaxoPerf for AI agents — MCP server connector + the maxoperf agent skill. Mirror of the private monorepo; do not edit directly.

Chrome breakpoint debugging MCP server for inspecting runtime values and stepping through code

Collective intelligence for agents. Stop your agents from guessing how to use the web. Let them know what to do. Think; hive mind. Let's make the open agent internet.

Web data for AI agents: scrape, crawl, search, deep research, site monitoring, browser automation

AI-powered browser automation MCP server for AI agents — 6 tools: web interaction, structured extraction, validation, multi-step tasks, memory enrichment.

Run a pinned Nocticas E2E gate in CI — fail the build on a red verdict. No false greens.

Val — the QA agent for vibecoders. Catch every bug, ship to green. GitHub Action that runs Val against your preview deployment URL and posts findings as a PR comment.

MCP server for DebugAI. Gives coding agents a debug_error tool that returns root cause, exact file and line, and ranked fixes as ready-to-apply edits. One-command setup, no key pasting.

Token-efficient browser agent for local LLMs — Playwright + accessibility tree + MarkGrab, MCP native.

Parse any user agent string into structured data: browser, OS, device type, engine, and bot detection in one call. -- x402 micropayment API + MCP server for AI agents

Uplink MCP server — automate a real, logged-in browser on your own device. npm: @uplink-code/mcp

AutoBrowser Approval Cloud public documentation modeled after the MiroFish docs pattern.

Paid remote MCP for AI browser purchase approval checks, spend thresholds, merchant evidence, risk notes, and audit receipts.

Chrome DevTools Session MCP public documentation modeled after the MiroFish docs pattern.

Paid remote MCP for Playwright selector risk checks, healing suggestions, action receipts, human approvals, and failure exports.

Playwriter Session Cloud public documentation modeled after the MiroFish docs pattern.

Docs, connection guide, and examples for the GameFork remote MCP server (io.gamefork/mcp) — publish, fork, and revive runnable browser games.

Allow LLMs to control a browser with Browserbase and Stagehand