MCP Library
Model Context Protocol servers that give AI agents real-world capabilities. Each entry includes tools, install commands, and transport details.
Machine-readable: GET /api/v1/mcps · ?capability=read_file
10 servers
Brave Search MCP
open sourcestdioGive agents real-time web search and local search via Brave's independent search index
Official MCP server for Brave Search API. Gives AI agents access to real-time web search results and local business/place search, using Brave's independent search index (not Google). Useful for agents that need up-to-date information without hallucination.
Fetch MCP
open sourcestdioLet agents fetch web pages and convert them to clean Markdown for processing
Official MCP server that gives agents the ability to fetch content from URLs and convert HTML pages to clean Markdown. Essential for agents that need to read documentation, extract content from URLs, or retrieve structured data from the web without a full browser.
Filesystem MCP
open sourcestdioOfficial Anthropic MCP server for reading and writing files on disk
The official Anthropic-maintained MCP server that gives AI agents secure, configurable access to the local filesystem. Supports reading, writing, listing, and searching files with access controls that restrict agents to specified directories.
GitHub MCP
open sourcestdioFull GitHub API access for agents: repos, issues, PRs, files, and search
Official MCP server for the GitHub API. Gives agents comprehensive access to GitHub: create and manage repositories, read and write files, open and merge pull requests, manage issues, search code, and interact with GitHub Actions. The most complete GitHub integration available for AI agents.
Memory MCP
open sourcestdioPersistent knowledge graph memory for AI agents across sessions
Official MCP server that gives AI agents a persistent, long-term memory store backed by a knowledge graph. Agents can create entities and relationships, store observations, and recall facts from previous sessions, enabling truly stateful AI agents that remember context over time.
Postgres MCP
open sourcestdioGive AI agents read access to any PostgreSQL database via natural language
An official MCP server that connects AI agents directly to a PostgreSQL database. Agents can list tables, inspect schemas, and run read-only SQL queries, enabling agents to answer data questions without a separate data pipeline.
SQLite MCP
open sourcestdioFull read and write access to a local SQLite database with schema inspection
Official MCP server for SQLite databases. Unlike the Postgres MCP (read-only), this server supports full read and write operations, agents can create tables, insert rows, run queries, and inspect schema. Ideal for agents that need lightweight persistent storage without an external database.
Puppeteer MCP
open sourcestdioBrowser automation for agents: navigate, click, screenshot, and scrape the web
Official MCP server that gives AI agents full browser automation via Puppeteer. Agents can navigate to URLs, interact with elements, fill forms, take screenshots, and execute JavaScript, enabling web scraping, UI testing, and browser-based automation without a headless browser setup.
Sequential Thinking MCP
open sourcestdioGive agents a structured dynamic reasoning scratchpad for complex problem solving
Official MCP server that provides a dynamic, structured scratchpad for agents to think through complex problems step by step. Enables multi-step reasoning with branching, revision, and hypothesis tracking, improving agent quality on hard analytical and planning tasks.
Slack MCP
open sourcestdioGive agents read and write access to Slack channels, messages, and threads
Official MCP server for Slack. Agents can list channels, read message history, post messages, reply in threads, add reactions, and look up user profiles, enabling agents to participate in team communication, monitor channels, and send notifications automatically.