"Kit" matching MCP servers:

Enables automation of Xiaohongshu (Little Red Book) operations including content publishing, searching, and interacting, with MCP protocol support for AI agent integration.

Enables Codex to delegate tasks to ChatGPT through a browser-backed Oracle connection, with project memory and a verified-result cache. It supports preview, delegation, result polling, and explicit acceptance without an API fallback.

Exposes BitBrowser automation to MCP hosts over stdio, letting agents check online status and inspect windows, tabs, and displays, and carry out authorized window creation, launch, close, delete, rename, URL browsing, tab actions, fingerprint randomization, grid/diagonal/adaptive arrangement, and Cookie or HTTP cache cleanup through the official local API. Write operations follow a prepare → execute → inspect flow with handoff artifacts built from tab snapshots.

Enables MCP-based diagnostics and automation of Tauri 2 WebView2 instances on Windows, supporting discovery, snapshots, screenshots, and input control without Chrome DevTools or remote debugging.

Enables coding agents to make structured local decisions, choose bounded browser actions with approvals, and run coding workflows such as file relevance, context pruning, diff screening, and completion checks through MCP or CLI.

An MCP Server that enables AI assistants to interact with your local browsers.

Enables AI assistants to inspect values, view errors, and capture canvas output from Observable Notebook Kit notebooks running in a web browser.

Enables LLM agents to read and navigate web pages efficiently with token budgets, extract structured data, and optionally control a browser with explicit user consent, remaining read-only by default.

Enables AI assistants to manage HelloFresh meal kit accounts by browsing menus, selecting recipes, and modifying delivery schedules. It supports updating dietary preferences, managing subscriptions, and rating past orders through Playwright-based browser automation.

Let AI agents use the Firefox, Zen, or LibreWolf browser you already have open, with your logins, cookies, and extensions intact. No Playwright, Chromium, or WebDriver. Agents work in background tabs while you keep browsing: read pages, click, fill forms, take screenshots, and capture network traffic. A WebExtension plus a local bridge lets several agents share one browser.

A lightweight MCP server for browser automation that gives AI agents navigate, screenshot, and extract tools via a single small binary, without needing Node.js or Playwright. It provides CDP-based browser control with efficient startup and resource usage.

An accessibility audit server that enforces WCAG 2.2 AA standards by running axe-core in a headless browser, providing violation details grouped by impact with fix hints.

Enables private browser automation for ChatGPT through Cloudflare Browser Run, with OAuth and owner-key authorization.

MCP server for lightweight web browsing: fetch and parse pages, extract structured content, run isolated JavaScript, and render to SVG/PNG/PDF with a minimal CDP interface.

MCP server that drives Cloudflare Kitesurf, an agent-first browser on Workers, enabling web page navigation, content extraction, screenshots, PDFs, accessibility trees, and JS evaluation from any MCP client without a local browser or API token.

Universal Search-First Knowledge Acquisition Plugin for LLMs. Enables real-time web search and deep page browsing via MCP or CLI. Zero-cost, privacy-first, supports DuckDuckGo, Bing, Google, Brave, Wikipedia, Arxiv, YouTube, Reddit and more.

Enables interaction with Google Maps through a dedicated browser session, supporting searches, directions, map views, and Street View without requiring the Google Maps Platform API.

Provides browser automation tools for coding agents—goto, browse, goal, page_state, and summary—so agents can drive real pages through auditable, gated decisions and record traces for offline replay.

Stealth browser automation over raw CDP, driving a real unmodified Chrome with no Playwright or Puppeteer in the stack. Returns numbered element refs from the accessibility tree so a model clicks ref 3 instead of guessing a CSS selector, and can attach to an already-running Chrome to reuse logged-in sessions.

Lets an AI assistant read and fill an Innoslate project through your own logged-in browser session, surveying the project, reading your documents and guides, and proposing a plan before writing entities, trace links, artifacts and reports after a dry run. It works on the free tier where API keys are refused, since every call goes through the browser instead.