Read, write, and manage files on the local filesystem with configurable access controls.
Persistent key-value memory store that lets AI models remember information across sessions.
Structured reasoning tool that helps models break down problems into sequential thought steps.
Fetch web pages, APIs, and raw URLs. Converts HTML to Markdown for cleaner AI consumption.
Search repos, read files, manage issues and PRs, and interact with the full GitHub API.
Search repos, manage issues, merge requests, and interact with the GitLab API.
Read local git repositories: log, diff, status, blame, and branch information.
Connect to PostgreSQL databases, run read-only queries, and explore schema.
Interact with local SQLite databases — query, create tables, and inspect schema.
Search, read, and manage files stored in Google Drive.
Search places, get directions, geocode addresses using the Google Maps Platform.
Read and send messages, list channels, and interact with Slack workspaces.
Search the web using the Brave Search API. Returns summarised results with sources.
Control a headless Chromium browser: navigate pages, click, fill forms, take screenshots.
Generate images using EverArt AI models directly from your AI assistant.
Query AWS Bedrock Knowledge Bases for RAG-powered retrieval from your own documents.
Browser automation with Playwright — navigate, click, extract content, and run tests.
Read and write notes in an Obsidian vault. Search, create, update, and link notes.
Create, read, and update pages and databases in your Notion workspace.
Create and manage Linear issues, projects, cycles, and teams from your AI assistant.
Query and manage Jira issues, projects, sprints, and users.
Read Figma files, extract components, inspect properties, and export assets.
Connect to MySQL databases, run read-only queries, and explore schema structure.
Interact with Redis — get, set, delete keys and inspect data structures.
Query MongoDB collections, aggregate data, and inspect documents.
List containers and images, run commands, inspect logs, and manage Docker resources.
Inspect pods, deployments, services, and namespaces in a Kubernetes cluster.
Manage Cloudflare Workers, KV, R2 storage, and D1 databases.
Deploy projects, manage environments, inspect logs, and control Vercel deployments.
Query Supabase databases, manage tables, edge functions, and storage buckets.
List customers, retrieve payments, inspect subscriptions, and read Stripe data.
Search the web using Perplexity AI for up-to-date answers with citations.
Search videos, read transcripts, and retrieve metadata from YouTube.
Search and control Spotify playback, read library, playlists, and track info.
Fetch top stories, comments, and user data from Hacker News.
Query PlanetScale MySQL-compatible databases and inspect branches and schema.
Query Turso (libSQL) edge databases with read-only SQL access.
This is an independent directory for informational purposes. All product names and trademarks are the property of their respective owners. CodeLint.Dev is not affiliated with, endorsed by, or sponsored by Anthropic, GitHub, or any other listed project. Archived servers are read-only reference copies; prefer actively maintained alternatives.
37 of 37 serversAbout
The Free MCP Servers directory lists 37 free and open-source Model Context Protocol (MCP) servers — including 16 official Anthropic-maintained servers and 21 community-built servers. MCP is an open standard that lets AI assistants like Claude connect to external tools and data sources: file systems, databases, APIs, browsers, and productivity apps. Each card links directly to the server's GitHub repository so you can install and configure it in minutes.
How to use
- 1 Browse by category (System, Database, Dev Tools, Web, Browser, etc.) or search by name or keyword.
- 2 Cards with an "Official" badge are maintained by Anthropic as reference implementations.
- 3 Click any card to open the GitHub repository with installation instructions.
- 4 Add a server to Claude Code by running: npx -y @anthropic-ai/claude-code mcp add <server-name>.
- What is the Model Context Protocol (MCP)?
- MCP is an open standard developed by Anthropic that defines a universal interface between AI assistants and external tools. It lets AI models like Claude securely read files, query databases, browse the web, call APIs, and interact with applications — replacing one-off integrations with a single, consistent protocol. MCP servers are lightweight processes that expose these capabilities.
- Are MCP servers free to use?
- All servers listed here are free and open-source. Some wrap third-party APIs (like Brave Search or Google Maps) that have their own free tiers and rate limits. The MCP server software itself is always free — you only pay for the underlying API usage if the service you connect to charges for it.
- How do I install an MCP server for Claude Code?
- Run npx -y @anthropic-ai/claude-code mcp add <server-name> in your terminal, replacing <server-name> with the server's npm package name (shown on each card). Claude Code will install and configure the server automatically. For servers not on npm, clone the GitHub repository and follow the manual installation instructions in the README.