Packages

Stability AI MCP server enabling AI agents to generate images via Stability AI's API — creating images from text prompts with Stable Diffusion models (SDXL, SD3, Stable Image), inpainting and outpainting, image-to-image transformation, and integrating Stability AI's generative image capabilities into agent-driven creative and content production workflows.

Data.gov MCP server collection enabling AI agents to access US government open data — searching federal datasets across agencies, retrieving dataset metadata and download URLs, querying CKAN-based data catalogs, and integrating open government data into agent-driven research, policy analysis, and data science workflows.

DeepSeek MCP server enabling AI agents to interact with DeepSeek language models — sending prompts to DeepSeek Chat and DeepSeek Coder models, executing code reasoning tasks, generating code completions, and integrating DeepSeek's cost-effective LLM capabilities into agent-driven coding assistance and AI inference workflows.

Official PlayCanvas Editor MCP server enabling AI agents to interact with the PlayCanvas 3D game engine editor — creating and modifying scene entities, managing components and assets, scripting game logic, and integrating PlayCanvas's WebGL-based game engine into agent-driven 3D content creation and game development workflows.

MCP server providing a live Python interpreter for AI agents. Enables agents to execute Python code, run scripts, evaluate expressions, and work with Python data in a runtime environment — supporting AI-assisted Python development, data analysis, and code verification with actual execution feedback.

SearXNG MCP server enabling AI agents to perform privacy-respecting web searches via SearXNG — a free, open-source metasearch engine — querying multiple search engines simultaneously, accessing search results without tracking, and integrating privacy-first web search into agent-driven research and information retrieval workflows using self-hosted or public SearXNG instances.

Official Sonatype MCP server providing AI agents with software composition analysis (SCA) capabilities — identifying vulnerabilities, license issues, and quality problems in open source dependencies.

Airtable's REST API for managing records in Airtable bases — a hybrid spreadsheet-database platform popular for structured data management without code.

Emacs MCP server enabling AI agents to interact with a running Emacs instance — executing Emacs Lisp commands, reading and modifying buffer contents, navigating files, controlling Org-mode, managing the editor state, and integrating AI assistance directly into Emacs editing workflows. Enables Claude or other AI agents to act as powerful Emacs coding assistants.

MCP server wrapping the Etherscan API — the leading Ethereum blockchain explorer. Enables AI agents to query Ethereum transaction history, wallet balances, smart contract ABIs, token transfers, gas prices, and on-chain analytics without running a full Ethereum node. Supports read-only blockchain data access for research and analytics agents.

A Python framework for building MCP servers and clients with automatic schema generation, validation, and transport management — used by 70% of MCP servers across all languages.

MCP server for GitHub Actions CI/CD platform. Enables AI agents to list, trigger, monitor, and manage GitHub Actions workflows programmatically. Agents can check workflow run status, trigger deployments, analyze build failures, and interact with GitHub's CI/CD pipeline through the GitHub Actions API.

MCP server enabling AI agents to interact with GitLab — reading and creating merge requests, managing issues, triggering CI/CD pipelines, accessing repository code, reviewing pipeline status, and interacting with GitLab's DevOps platform through MCP tool calls.

MCP server bridging AI assistants with KiCAD for PCB design automation. Provides 64 tools for component placement, routing, schematic editing, DRC checks, Gerber export, and JLCPCB parts integration with 2.5M+ components. Uses a smart router pattern to reduce context by 70%.

MCP Memory Keeper server providing persistent agent memory via SQLite — storing and retrieving memories with tags and categories, searching memories by content, managing memory collections, and integrating persistent knowledge storage into agent workflows that need to remember facts, decisions, and context across conversations.

PDF Extraction MCP server enabling AI agents to extract text and content from PDF files — reading PDF documents, extracting text by page, parsing structured content, and integrating PDF processing into agent-driven document analysis, RAG pipeline, and content extraction workflows.

Prometheus MCP server enabling AI agents to query Prometheus monitoring data — executing PromQL queries, retrieving metric metadata, inspecting alert rules and firing alerts, and integrating Prometheus's open-source metrics monitoring into agent-driven SRE, incident response, and infrastructure observability workflows.

Community MCP server providing AI assistants real-time access to official SEC EDGAR filings and financial data. Enables agents to search and retrieve 10-K, 10-Q, 8-K filings, earnings data, and SEC-registered company information from the SEC's public EDGAR database — essential for financial research and investment analysis workflows.

MCP server implementing sequential thinking and structured reasoning capabilities for AI agents. Provides tools for breaking complex problems into sequential steps, maintaining reasoning chains, tracking thought progressions, and supporting systematic problem decomposition — enhancing AI agent reasoning quality.

Unla is a lightweight MCP gateway written in Go that converts existing REST APIs and MCP servers into MCP-compliant SSE or Streamable HTTP endpoints with zero code changes, providing multi-tenant management, OAuth pre-auth, JWT support, and hot-reload configuration.