DEDUPED Hot AI repos. Real use cases. What people build. Get the Monday 5

BEST AI REPOS BY TASK

Best open-source AI repos for MCP servers (2026)

6 repos compared · updated 2026-10-27

What are the best open-source MCP servers?

The best open-source AI repos for MCP servers right now are GitHub MCP Server, Playwright MCP and Chrome DevTools MCP. The most useful first MCP servers connect your agent to tools you already use: GitHub MCP Server for repos, issues and pull requests, Playwright MCP or Chrome DevTools MCP for the browser, and Firecrawl for the web.

Comparison

RepoWhat it doesStarsStart here
GitHub MCP Server
github/github-mcp-server · Go		The GitHub MCP Server is GitHub's official server that connects AI tools to GitHub.33,932Connect your agent to GitHub with the hosted MCP server Beginner, 10 minutes
Playwright MCP
microsoft/playwright-mcp · TypeScript		Playwright MCP is Microsoft's MCP server for browser automation with Playwright.38,601Give Claude Code or Codex a real browser in one command Beginner, 5 minutes
Chrome DevTools MCP
ChromeDevTools/chrome-devtools-mcp · TypeScript		Chrome DevTools for agents (chrome-devtools-mcp) lets a coding agent such as Claude, Cursor, Copilot or Antigravity control and inspect a live Chrome browser, as an MCP server or as a CLI.54,099Let your coding agent debug a page in a real Chrome Beginner, 10 minutes
Firecrawl
firecrawl/firecrawl · TypeScript		Firecrawl is an API that searches, scrapes, crawls and interacts with the web and returns clean markdown, structured JSON or screenshots for AI agents.195,071Turn any web page into clean markdown for an LLM Beginner, 10 minutes
MCP for Blender
ahujasid/mcp-for-blender · Python		MCP for Blender (formerly blender-mcp) connects Blender to any LLM through the Model Context Protocol.32,422Build a 3D scene in Blender from a text prompt Beginner, 20 minutes
Context Mode
mksglu/context-mode · TypeScript		Context Mode is an MCP server and hook set for AI coding agents that keeps raw tool output out of the context window, persists session memory in SQLite, and routes work across many clients.31,762Keep large tool output out of Claude Code's context Beginner, 15 minutes

When to pick which

GitHub MCP Server github/github-mcp-server

The GitHub MCP Server is GitHub's official server that connects AI tools to GitHub. Agents can browse and search code, manage issues and pull requests, watch GitHub Actions runs and analyze failures, review Dependabot alerts and security findings, and work with discussions and notifications. You can use the remote server hosted by GitHub or run it locally in Docker with a personal access token, and turn toolsets on or off.

Best for: Developers and maintainers who want their agent or chatbot to work with issues, PRs and CI on GitHub directly.

Needs: A GitHub account; An MCP host such as VS Code, Claude Code or Cursor; For local use: Docker and a GitHub personal access token

Start here: Connect your agent to GitHub with the hosted MCP server Beginner, 10 minutes, works with claude-code, cursor, codex

  1. Add a server named github of type http with the URL https://api.githubcopilot.com/mcp/ to your host's MCP config
  2. Sign in with OAuth, or pass a personal access token in the Authorization header
  3. Ask the agent about open issues, recent commits or a PR

All 3 use casesGitHub

Playwright MCP microsoft/playwright-mcp

Playwright MCP is Microsoft's MCP server for browser automation with Playwright. It lets an LLM work with web pages through structured accessibility snapshots instead of screenshots, so no vision model is needed. It installs with one npx command in Claude Code, Codex, Cursor, VS Code, Amp and other clients, and supports Chrome, Firefox, WebKit and Edge. The README suggests the Playwright CLI with skills for token-heavy coding agent work and MCP for exploratory, stateful agent loops.

Best for: Developers and QA engineers who want an agent to click through a web app, test flows or collect data in a real browser.

Needs: Node.js; An MCP client such as Claude Code, Codex, Cursor or VS Code

Start here: Give Claude Code or Codex a real browser in one command Beginner, 5 minutes, works with claude-code, codex, cursor

  1. Claude Code: claude mcp add playwright npx @playwright/mcp@latest
  2. Codex: codex mcp add playwright npx "@playwright/mcp@latest"
  3. Other clients: add an mcpServers entry named playwright that runs npx with @playwright/mcp@latest
  4. Ask the agent to open your app and walk through a flow

All 3 use casesGitHub

Chrome DevTools MCP ChromeDevTools/chrome-devtools-mcp

Chrome DevTools for agents (chrome-devtools-mcp) lets a coding agent such as Claude, Cursor, Copilot or Antigravity control and inspect a live Chrome browser, as an MCP server or as a CLI. The agent can record performance traces and get insights, read network requests and console messages with source-mapped stack traces, take screenshots, and automate pages with Puppeteer. The README warns that it exposes the browser's content to the MCP client, so avoid sensitive sites in that browser.

Best for: Web developers who want their coding agent to see what the page actually does in Chrome instead of guessing from source.

Needs: Node.js and npm; Chrome; An MCP client or a coding agent that can run shell commands

Start here: Let your coding agent debug a page in a real Chrome Beginner, 10 minutes, works with claude-code, cursor, gemini-cli

  1. Add an mcpServers entry named chrome-devtools that runs npx with chrome-devtools-mcp@latest
  2. Restart your MCP client so the tools load
  3. Ask the agent to open your local app and report console messages and failed network requests
  4. Have it fix the cause and take a screenshot to confirm

All 3 use casesGitHub

Firecrawl firecrawl/firecrawl

Firecrawl is an API that searches, scrapes, crawls and interacts with the web and returns clean markdown, structured JSON or screenshots for AI agents. It handles JavaScript-heavy pages, proxies and rate limits, parses PDFs and DOCX, can click and type before extracting, and plugs into agents as a skill or an MCP server. It is open source and also offered as a hosted service with an API key.

Best for: Developers building AI agents, RAG pipelines or research tools that need clean page content instead of raw HTML.

Needs: A Firecrawl API key from the hosted service, or a self-hosted instance; Python, Node.js, or curl; or an MCP client such as Claude Code or Cursor

Start here: Turn any web page into clean markdown for an LLM Beginner, 10 minutes

  1. Get an API key at Firecrawl.dev
  2. Python: from firecrawl import Firecrawl, create app = Firecrawl(api_key="fc-YOUR_API_KEY"), then result = app.scrape('firecrawl.dev')
  3. Or from the CLI: firecrawl https://firecrawl.dev --only-main-content
  4. Ask for other formats such as structured JSON or screenshots when you need them

All 3 use casesGitHub

MCP for Blender ahujasid/mcp-for-blender

MCP for Blender (formerly blender-mcp) connects Blender to any LLM through the Model Context Protocol. A Blender addon runs a socket server and an MCP server connects it to Claude Desktop, Claude Code, Codex, Cursor, VS Code and others, so the AI can create and modify objects, apply materials, check its work with multi-angle views and renders, run Blender Python and pull in Poly Haven, Sketchfab and Poly Pizza assets or AI-generated 3D models.

Best for: 3D artists, game developers and hobbyists who want to build and edit Blender scenes by describing them in plain language.

Needs: Blender; uv (the setup installs it if needed; use the official installer, not pip install uv); An MCP client such as Claude Desktop, Claude Code, Codex or Cursor; Optional: API access for AI model generators, or the Premium plan

Start here: Build a 3D scene in Blender from a text prompt Beginner, 20 minutes, works with claude-code, codex, cursor

  1. On macOS or Linux run curl -LsSf https://www.mcp-for-blender.com/install.sh | sh (or uvx mcp-for-blender setup if you already have uv)
  2. Pick the MCP clients to set up; it adds a blender entry and saves a .bak of each config
  3. Fully quit and reopen your AI app, then open Blender; the addon starts its server
  4. Ask the AI to build a scene, for example a low-poly campsite at dusk

All 3 use casesGitHub

Context Mode mksglu/context-mode

Context Mode is an MCP server and hook set for AI coding agents that keeps raw tool output out of the context window, persists session memory in SQLite, and routes work across many clients.

Best for: Developers who use AI coding agents on large codebases and notice their context window filling up during long sessions.

Needs: Claude Code v1.0.33+ for the plugin install, or Node.js >= 22.5 (or Bun) for the Gemini CLI setup; An existing AI coding agent session in a project you want to work on

Start here: Keep large tool output out of Claude Code's context Beginner, 15 minutes, works with claude-code

  1. Run /plugin marketplace add mksglu/context-mode in Claude Code.
  2. Run /plugin install context-mode@context-mode.
  3. Restart Claude Code or run /reload-plugins.
  4. Run /context-mode:ctx-doctor and confirm all checks show [x].

All 5 use casesGitHub

Use cases for MCP servers

Questions

What are the best open-source MCP servers?

The best open-source AI repos for MCP servers right now are GitHub MCP Server, Playwright MCP and Chrome DevTools MCP. The most useful first MCP servers connect your agent to tools you already use: GitHub MCP Server for repos, issues and pull requests, Playwright MCP or Chrome DevTools MCP for the browser, and Firecrawl for the web.

Which repo for MCP servers is easiest to start with?

GitHub MCP Server: its starter use case "Connect your agent to GitHub with the hosted MCP server" is rated Beginner and takes about 10 minutes.

How do I get started with GitHub MCP Server?

Add a server named github of type http with the URL https://api.githubcopilot.com/mcp/ to your host's MCP config; Sign in with OAuth, or pass a personal access token in the Authorization header; Ask the agent about open issues, recent commits or a PR. Full steps and a copy-paste agent prompt: https://dedupedai.com/use-cases/github-github-mcp-server/connect-your-agent-to-github-remotely/

Are these MCP servers repos free?

They are open-source projects on GitHub (check each license; some, like fair-code projects, limit commercial use). Some also sell a hosted version or need a paid model API key; the requirements for each repo are listed on this page.

How is this list of MCP servers repos chosen?

Deduped tracks the hottest AI repos on GitHub and writes step-by-step use cases from each README. For this page we picked the repos that do this job, ordered by fit and then GitHub stars. Repos we only inferred from a short README, or that we do not index, are left out. The page is rebuilt with the site; stars are as of the last build in 2026.

More answers

For AI agents: Markdown version · JSON · MCP server · llms.txt

Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer