Packages

MCP server that replicates Claude Code capabilities — file read/write/edit, grep/search, shell command execution, Jupyter notebook support, sub-agent delegation, and todo tracking. Supports any LiteLLM-compatible model for agent operations, enabling Claude Code-like workflows in any MCP client.

Memory MCP Server providing persistent memory storage for AI agents — storing facts, preferences, conversation summaries, and contextual information across sessions, enabling agents to remember user preferences, past conversations, and learned information beyond the context window through structured memory operations.

MetaMCP is an MCP proxy and aggregator server enabling AI agents to manage and route across multiple MCP servers from a single endpoint — dynamically discovering available MCP tools, proxying requests to underlying MCP servers, providing a unified interface for multi-MCP environments, and enabling flexible MCP server composition without modifying agent configuration.

MCP server using PBIXRay to analyze Power BI (.pbix) files. Enables AI agents to inspect Power BI report internals — DAX measures, data models, relationships, data sources, and metadata — without requiring Power BI Desktop. Useful for Power BI governance, auditing, and documentation automation.

An MCP server by Microsoft that enables LLMs to automate web browsers using structured accessibility snapshots instead of screenshots, providing 30+ tools for navigation, interaction, and testing.

Sandbox MCP server providing secure, isolated code execution environments for AI agents — running code in sandboxed containers to prevent unauthorized access to the host system, enabling agents to safely execute untrusted or generated code without security risk to the host machine.

Official Tabby MCP server enabling AI agents to interact with Tabby — the self-hosted, open-source AI coding assistant — querying Tabby for code completions, accessing repository context indexed by Tabby, retrieving code search results, and integrating Tabby's local code intelligence into agent-driven development workflows without sending code to external services.

Lightweight MCP server for task management by flesler — a known open-source contributor. Enables AI agents to create, read, update, and delete tasks with priority and status tracking — providing a simple persistent task store for AI-driven productivity workflows without external service dependencies.

MCPGen — a CLI code generator for creating MCP server scaffolding — generating boilerplate code for new MCP servers in TypeScript or Python, scaffolding tool and resource definitions, creating project structure with proper configuration, and accelerating MCP server development by automating the repetitive setup of new MCP server projects.

MCP server providing AI agents with access to AlphaFold Protein Structure Database — DeepMind/EBI's groundbreaking database of predicted protein structures for 200M+ proteins. Enables agents to retrieve protein structure predictions by UniProt accession, access confidence scores, download structure files (PDB/mmCIF), and integrate protein structure data into biological research workflows.

Official Crawlbase MCP server enabling AI agents to use Crawlbase's web scraping and crawling API — fetching web pages through residential/datacenter proxies, bypassing anti-bot protection, extracting structured data from websites, and integrating web data collection into agent-driven research and automation workflows.

Official HashiCorp Vault MCP server enabling AI agents to interact with Vault for secrets management — reading/writing secrets, managing leases, querying PKI, and interacting with Vault's secrets engines.

MCP Auth Python SDK providing authentication middleware and utilities for building secure MCP servers — implementing OAuth 2.0, API key validation, and other auth flows in Python MCP server applications, enabling developers to add proper authentication to their custom MCP servers with minimal boilerplate.

Overture is an open-source MCP server that intercepts AI coding agent planning phases and renders execution plans as interactive visual flowcharts in a web UI before code is written, allowing developers to inspect, enrich, and approve plans.

MCP server providing access to F# Interactive (FSI) — the F# REPL and scripting environment. Enables AI agents to evaluate F# expressions and scripts in a live FSI session, inspect values, and interact with .NET/F# code dynamically — enabling AI-assisted F# development with real evaluation capabilities.

Figma MCP server enabling AI agents to interact with Figma design files — reading file structure and components, inspecting design tokens and styles, extracting layout and element properties, navigating frames and pages, and integrating Figma's design platform into agent-driven design-to-code, design audit, and documentation workflows.

Hugging Face MCP server enabling AI agents to interact with the Hugging Face ecosystem — running model inference via Inference API, searching and discovering models from the Hub, accessing datasets, and integrating the world's largest open model repository into agent workflows.

Local Kanban board MCP server enabling AI agents to manage kanban-style task boards — creating and moving cards between columns (To Do, In Progress, Done), managing boards and lists, tracking task progress, and integrating kanban workflow management into agent-driven project coordination. Provides lightweight local kanban without requiring external services.

MCP configuration management tool that helps developers manage and configure multiple MCP servers across different AI clients. Provides a unified interface for adding, removing, and managing MCP server configurations in Claude Desktop, Cursor, and other MCP-compatible clients without manually editing JSON config files.

Symfony bundle for building MCP servers in PHP — providing a framework for PHP/Symfony developers to create custom MCP servers using Symfony's component system. Enables backend PHP applications to expose functionality as MCP tools, making it easy for Symfony teams to build MCP integrations for their existing PHP applications.