Browse the catalog

Provides tools for making small, targeted edits to files using stream editors like sed and awk, enabling efficient modifications without full file replacement.

Enables MCP-capable clients to inspect and manage Keycloak realms, users, clients, roles, and groups with layered security modes, realm allowlisting, protected realms, delete gating, dry-run, and audit logging.

Enables AI assistants to search, create, and manage Notion workspace content including pages and databases. It supports advanced database querying, page updates, and content organization through natural language conversations.

Provides complete integration with Coolify's self-hosted PaaS platform through its REST API, enabling management of applications, databases, services, servers, and deployments via natural language commands. Built by Pavel Sukhachev with TypeScript and comprehensive test coverage, this server exposes over 40 tools for operations like creating Git-based applications, setting up databases (PostgreSQL, MySQL, MongoDB, Redis), managing environment variables, executing commands in containers, and monitoring deployment logs. Includes pre-built prompts for common workflows like deploying applications from repositories, setting up SSL certificates, scaling resources, and troubleshooting failed deployments, making it valuable for developers who want to manage their Coolify infrastructure through AI assistants without switching to the web interface.

Comment directives in, compact AI context out — a drift-aware code graph for JS/TS that gives Claude, Codex & other assistants a small query pack instead of the whole tree. CLI + MCP server.

Enables Claude to read and analyze PDF documents with automatic OCR processing for scanned files. Features intelligent text extraction, caching for performance, and secure file access with search capabilities.

SlowMCP is a modern framework for building, testing, and shipping MCP servers, written in CoffeeScript. It includes a protocol-backed test harness and works over Streamable HTTP.

Brex MCP Server provides a secure bridge between AI agents and the Brex financial platform, enabling access to account information, transactions, expenses, and budget resources. Developed by Dennison Bertram, this TypeScript implementation exposes Brex API data through standardized resource handlers and tools following the MCP specification. The server implements read-only operations for sensitive financial resources, proper error handling, and rate limiting for API requests. It's particularly valuable for financial analysts using AI tools to retrieve and analyze Brex financial data without requiring direct API access, supporting use cases like expense analysis, budget monitoring, and transaction categorization.

GraphRAG MCP provides a hybrid retrieval system that combines Neo4j graph database and Qdrant vector database capabilities for powerful document search and context expansion. Developed by Riley Lemm, this server implementation enables semantic search through document embeddings, graph-based context expansion following relationships, and hybrid search combining both approaches. The server exposes tools for querying documentation and resources for accessing database schema information, making it particularly valuable for applications requiring both semantic relevance and structural context in document retrieval, such as technical documentation systems, knowledge bases, or any application needing contextually aware information retrieval.

MCP server on Cloudflare Workers with OAuth login, enabling remote tool calls like math operations from MCP clients.

invinoveritas SDK — Lightning-native AI agent platform. Pay-per-use reasoning, marketplace buys, Lightning withdrawals behind a single Bearer key.

Twitter/X MCP server for the API v2, enabling users to manage profiles, posts, likes, retweets, and follows through natural language using a stateless Streamable HTTP transport with OAuth user access token authentication.

Backend services for Options Sentinel, providing AI agent orchestration, market analysis, risk validation, options strategy generation, and Alpaca MCP-based paper-trading execution.

Brings DuckDuckGo search (web, news, images, videos) to any MCP-compatible AI client without requiring an API key.

CLI-native multi-agent collaboration for Claude Code: hot-pluggable role yamls, persistent memory MCP, step-level trace replay.

A Model Context Protocol server that provides tools for querying, updating, and exploring Databricks SQL Analytics tables.

Converts web pages to Markdown with tiered fetching to minimize LLM context usage, enabling efficient extraction of specific sections.

A terminal-based MCP server that provides tools like web search, URL fetching, note management, and calculations, and connects them to a LangGraph ReAct agent for multi-step tasks.

A TypeScript-based MCP server that implements a simple notes system, allowing users to create, access, and generate summaries of notes via Claude Desktop integration.

CodeLedger provides AI coding agents with direct access to repository memory through three MCP tools for querying verified patterns, retrieving task-specific context bundles, and recording interaction outcomes. Operates as a local CLI with stdio MCP transport, processing repository data without external transmission. Compatible with Claude Code, Cursor, and Windsurf.