Browse the catalog

SEC ADV / IAPD MCP — SEC Investment Adviser Public Disclosure (Form ADV) lookup

A basic starter project for building Model Context Protocol (MCP) servers that enables standardized interactions between AI systems and various data sources through secure, controlled tool implementations.

A community-curated list of MCP (Model Context Protocol) servers. PRs welcome! Browse AI tools, integrations, and developer resources.

The MCP-CEP server provides a simple interface for querying Brazilian postal codes (CEPs) via the ViaCEP API. Developed by Wendell Barreto, this implementation runs as a stdio-based server that allows AI assistants to retrieve address information directly from Brazil's postal code database. The server is built using Python with the httpx library for API requests and the MCP framework for standardized communication. It's particularly useful for applications that need to validate or expand Brazilian addresses based on postal codes without requiring users to leave their conversation context.

This MCP server, developed to access BART (Bay Area Rapid Transit) departures data, provides real-time information on train schedules and station updates. Built with TypeScript and leveraging the Model Context Protocol SDK, it offers a simple interface for querying BART's API and retrieving structured departure information. The implementation stands out by focusing specifically on BART data, making it ideal for applications or AI assistants needing up-to-date public transit information for the San Francisco Bay Area. It's particularly useful for commuters, travel planners, or local businesses looking to integrate real-time BART data into their services or decision-making processes.

Apricot MCP public description

ArcGIS Pro MCP server by nicogis that enables AI assistants to interact with ArcGIS Pro desktop GIS software through a bridge architecture consisting of a .NET add-in running inside ArcGIS Pro and a separate MCP server process that communicates via inter-process communication. The implementation uses named pipes or similar IPC mechanisms to allow the MCP server to execute GIS operations, access map data, and control ArcGIS Pro functionality from external applications while maintaining the full desktop GIS environment. Built with C# for the ArcGIS Pro add-in component and .NET for the MCP server, this architecture enables AI assistants to perform spatial analysis, map creation, and GIS workflows without requiring direct API access, making it valuable for geospatial professionals who need conversational access to desktop GIS capabilities and automated spatial data processing tasks.

MCP server for Kubernetes FinOps that finds over-provisioned workloads, proposes right-sizing, and enforces safety guardrails with dry-run by default and audit logging.

Connect AI agents to your test results, insights, and targets. Query test runs, failures, flaky tests, and regressions across frameworks including Playwright, Jest, Pytest, Cypress and more.

This Azure Table Storage MCP server, developed by Christia Pedersen, provides AI assistants with a standardized interface to interact with Azure Table Storage. Built using TypeScript and leveraging the @azure/data-tables library, it offers capabilities for storing, retrieving, and querying structured data in the cloud. The server abstracts the complexities of Azure Table Storage operations, handling authentication, data serialization, and query optimization. It's particularly valuable for developers and AI systems working with large-scale, semi-structured data, enabling use cases such as log data storage, user profile management, and IoT telemetry processing within the Azure ecosystem.

Enables natural language analysis of network packet captures, including protocol detection, flow analysis, and security threat identification, integrated with AI assistants.

A demonstration MCP server for local development and testing with Claude Desktop on WSL. Provides basic utility tools including greeting messages, echo functionality, and server information retrieval.

This project demonstrates a decoupled real-time agent architecture that connects LangGraph agents to remote tools served by custom MCP (Modular Command Protocol) servers. The architecture enables a flexible and scalable multi-agent system where each tool can be hosted independently (via SSE or STDIO), offering modularity and cloud-deployable execut

Enables playing and observing a running Unreal Engine 5 game through virtual gamepad, keyboard, and mouse input, plus screen capture, log streaming, and UMG UI inspection and control.

Provides verified statistical inference and hypothesis testing tools, including t-tests, effect sizes, power analysis, and multiple comparisons correction, with assumption checks and citations.

Enables Claude to interact with cognee for cognitive augmentation, supporting local or Smithery installation.

A hosted MCP server that gives AI agents live, self-custodial access to the Bittensor blockchain, enabling queries and on-chain actions via any MCP client.

Enables AI agents to get real home & auto insurance quotes and start binding through a network of licensed independent agencies.

Enables AI agents to autonomously provision, pay for, and manage S3-compatible storage and vector databases without human intervention.

Connects Claude and other MCP-compatible AI assistants to the AI Bounty Board, enabling natural language discovery, claiming, and submission of bounties.