Discover our curated collection of MCP servers for learning & documentation. Browse 1660servers and find the perfect MCPs for your needs.
Facilitates the creation of mathematical diagrams using natural language through Penrose's domain-specific languages.
Empowers AI assistants with comprehensive, up-to-date Minecraft modding documentation for Fabric and NeoForge.
Provides intelligent code analysis, validation, and documentation for Zig, leveraging a fine-tuned LLM and the official Zig compiler.
Provides local-first, hybrid semantic and keyword search for various document types and Confluence pages through the Model Context Protocol.
Reduces medical hallucinations and enhances clinical accuracy in radiology multimodal large language models.
Enables semantic search of user-provided ETABS documentation using local embeddings and a Model Context Protocol (MCP) server.
Facilitates a simple and stable connection between Anki and large language models.
Indexes and searches local project documentation efficiently for coding assistants and development tools.
Automates Cole Medin's context engineering methodology to generate high-quality Product Requirements Prompts (PRPs) through AI-assisted workflows.
Orchestrates expert AI agents as standard Model Context Protocol (MCP) tools, enabling host environments to access advanced task processing capabilities through a unified interface.
Guides aspiring programmers through Python fundamentals using an engaging space adventure narrative.
Automates intelligent documentation deployment for open-source projects through deep repository analysis and static site generator recommendations.
Enables users to enjoy the lateral thinking game 'Haiguitang' (also known as Situation Puzzles) alone by utilizing a large language model as the game master.
Enables AI agents to interact with the Open eClass platform for course information and basic operations.
Enables AI assistants to interact with Confluence content through a standardized interface.
Aggregates complete developer documentation for WeCom and Feishu open platforms, enabling rapid retrieval and review of interface documents within AI programming assistants.
Provides AI models with comprehensive access to the Semantic Scholar Academic Graph API for intelligent literature search and analysis.
Enables checking Mathematica code via a local Mathematica installation.
Exposes PubNub SDK documentation and API resources to LLM-powered tools, enhancing AI agent interaction with PubNub.
Indexes technical documentation on a local vector database and integrates with AI IDEs for semantic search.
Scroll for more results...