mcp-documentation-server

MCP Documentation Server - Bridge the AI Knowledge Gap. ✨ Features: Document management • AI-powered semantic search • File uploads • Smart chunking • Multilingual support • Zero-setup 🎯 Perfect for: New frameworks • API docs • Internal guides

Stars: 201

Visit

The mcp-documentation-server is a lightweight server application designed to serve documentation files for projects. It provides a simple and efficient way to host and access project documentation, making it easy for team members and stakeholders to find and reference important information. The server supports various file formats, such as markdown and HTML, and allows for easy navigation through the documentation. With mcp-documentation-server, teams can streamline their documentation process and ensure that project information is easily accessible to all involved parties.

README:

MCP Documentation Server

A TypeScript-based Model Context Protocol (MCP) server that provides local-first document management and semantic search using embeddings. The server exposes a collection of MCP tools and is optimized for performance with on-disk persistence, an in-memory index, and caching.

Demo Video

Core capabilities

O(1) Document lookup and keyword index through DocumentIndex for fast chunk and document retrieval.
LRU EmbeddingCache to avoid recomputing embeddings and speed up repeated queries.
Parallel chunking and batch processing to accelerate ingestion of large documents.
Streaming file reader to process large files without high memory usage.
Chunk-based semantic search with context-window retrieval to gather surrounding chunks for better LLM answers.
Local-only storage: no external database required. All data resides in ~/.mcp-documentation-server/.

Quick Start

Install and run

Run directly with npx (recommended):

npx @andrea9293/mcp-documentation-server

Configure an MCP client

Example configuration for an MCP client (e.g., Claude Desktop):

{
  "mcpServers": {
    "documentation": {
      "command": "npx",
      "args": [
        "-y",
        "@andrea9293/mcp-documentation-server"
      ],
      "env": {
        "MCP_EMBEDDING_MODEL": "Xenova/all-MiniLM-L6-v2"
      }
    }
  }
}

Basic workflow

Add documents using the add_document tool or by placing .txt, .md, or .pdf files into the uploads folder and calling process_uploads.
Search documents with search_documents to get ranked chunk hits.
Use get_context_window to fetch neighboring chunks and provide LLMs with richer context.

Features

Document management: add, list, retrieve, delete documents and metadata.
Semantic search: chunk-level search using embeddings plus an in-memory keyword index.
DocumentIndex: constant-time lookups for documents and chunks; supports deduplication and persisted index file.
EmbeddingCache: configurable LRU cache for embedding vectors to reduce recomputation and speed repeated requests.
Parallel and batch chunking: ingestion is parallelized for large documents to improve throughput.
Streaming file processing: large files are processed in a streaming manner to avoid excessive memory usage.
Context window retrieval: fetch N chunks before/after a hit to assemble full context for LLM prompts.
Local-first persistence: documents and index are stored as JSON files under the user's data directory.

Exposed MCP tools

The server exposes several tools (validated with Zod schemas) for document lifecycle and search:

add_document — Add a document (title, content, metadata)
list_documents — List stored documents and metadata
get_document — Retrieve a full document by id
delete_document — Remove a document and its chunks
process_uploads — Convert files in uploads folder into documents (chunking + embeddings)
get_uploads_path — Returns the absolute uploads folder path
list_uploads_files — Lists files in uploads folder
search_documents — Semantic search within a document (returns chunk hits and LLM hint)
get_context_window — Return a window of chunks around a target chunk index

Configuration & environment variables

Configure behavior via environment variables. Important options:

MCP_EMBEDDING_MODEL — embedding model name (default: Xenova/all-MiniLM-L6-v2). Changing the model requires re-adding documents. (all feature extraction xenova models are here).
MCP_INDEXING_ENABLED — enable/disable the DocumentIndex (true/false). Default: true.
MCP_CACHE_SIZE — LRU embedding cache size (integer). Default: 1000.
MCP_PARALLEL_ENABLED — enable parallel chunking (true/false). Default: true.
MCP_MAX_WORKERS — number of parallel workers for chunking/indexing. Default: 4.
MCP_STREAMING_ENABLED — enable streaming reads for large files. Default: true.
MCP_STREAM_CHUNK_SIZE — streaming buffer size in bytes. Default: 65536 (64KB).
MCP_STREAM_FILE_SIZE_LIMIT — threshold (bytes) to switch to streaming path. Default: 10485760 (10MB).

Example .env (defaults applied when variables are not set):

MCP_INDEXING_ENABLED=true          # Enable O(1) indexing (default: true)
MCP_CACHE_SIZE=1000                # LRU cache size (default: 1000)
MCP_PARALLEL_ENABLED=true          # Enable parallel processing (default: true)
MCP_MAX_WORKERS=4                  # Parallel worker count (default: 4)
MCP_STREAMING_ENABLED=true         # Enable streaming (default: true)
MCP_STREAM_CHUNK_SIZE=65536        # Stream chunk size (default: 64KB)
MCP_STREAM_FILE_SIZE_LIMIT=10485760 # Streaming threshold (default: 10MB)

Default storage layout (data directory):

~/.mcp-documentation-server/
├── data/      # Document JSON files
└── uploads/   # Drop files (.txt, .md, .pdf) to import

Usage examples

Add a document via MCP tool:

{
  "tool": "add_document",
  "arguments": {
    "title": "Python Basics",
    "content": "Python is a high-level programming language...",
    "metadata": {
      "category": "programming",
      "tags": ["python", "tutorial"]
    }
  }
}

Search a document:

{
  "tool": "search_documents",
  "arguments": {
    "document_id": "doc-123",
    "query": "variable assignment",
    "limit": 5
  }
}

Fetch context window:

{
  "tool": "get_context_window",
  "arguments": {
    "document_id": "doc-123",
    "chunk_index": 5,
    "before": 2,
    "after": 2
  }
}

Performance and operational notes

Embedding models are downloaded on first use; some models require several hundred MB of downloads.
The DocumentIndex persists an index file and can be rebuilt if necessary.
The EmbeddingCache can be warmed by calling process_uploads, issuing curated queries, or using a preload API when available.

Embedding Models

Set via MCP_EMBEDDING_MODEL environment variable:

Xenova/all-MiniLM-L6-v2 (default) - Fast, good quality (384 dimensions)
Xenova/paraphrase-multilingual-mpnet-base-v2 (recommended) - Best quality, multilingual (768 dimensions)

The system automatically manages the correct embedding dimension for each model. Embedding providers expose their dimension via getDimensions().

⚠️ Important: Changing models requires re-adding all documents as embeddings are incompatible.

Development

git clone https://github.com/andrea9293/mcp-documentation-server.git

cd mcp-documentation-server

npm run dev

npm run build

npm run inspect

Contributing

Fork the repository
Create a feature branch: git checkout -b feature/name
Follow Conventional Commits for messages
Open a pull request

License

MIT - see LICENSE file

Support

Star History

Built with FastMCP and TypeScript 🚀

For Tasks:

Click tags to check more tools for each tasks

manage documentation host project files share information access project details navigate documentation

For Jobs:

software developer technical writer project manager system administrator web developer

Alternative AI tools for mcp-documentation-server

Similar Open Source Tools

mcp-documentation-server

github

: 201

docs-mcp-server

The docs-mcp-server repository contains the server-side code for the documentation management system. It provides functionalities for managing, storing, and retrieving documentation files. Users can upload, update, and delete documents through the server. The server also supports user authentication and authorization to ensure secure access to the documentation system. Additionally, the server includes APIs for integrating with other systems and tools, making it a versatile solution for managing documentation in various projects and organizations.

github

: 558

mcp-server-mysql

The MCP Server for MySQL based on NodeJS is a Model Context Protocol server that provides access to MySQL databases. It enables users to inspect database schemas and execute SQL queries. The server offers tools for executing SQL queries, providing comprehensive database information, security features like SQL injection prevention, performance optimizations, monitoring, and debugging capabilities. Users can configure the server using environment variables and advanced options. The server supports multi-DB mode, schema-specific permissions, and includes troubleshooting guidelines for common issues. Contributions are welcome, and the project roadmap includes enhancing query capabilities, security features, performance optimizations, monitoring, and expanding schema information.

github

: 768

activepieces

Activepieces is an open source replacement for Zapier, designed to be extensible through a type-safe pieces framework written in Typescript. It features a user-friendly Workflow Builder with support for Branches, Loops, and Drag and Drop. Activepieces integrates with Google Sheets, OpenAI, Discord, and RSS, along with 80+ other integrations. The list of supported integrations continues to grow rapidly, thanks to valuable contributions from the community. Activepieces is an open ecosystem; all piece source code is available in the repository, and they are versioned and published directly to npmjs.com upon contributions. If you cannot find a specific piece on the pieces roadmap, please submit a request by visiting the following link: Request Piece Alternatively, if you are a developer, you can quickly build your own piece using our TypeScript framework. For guidance, please refer to the following guide: Contributor's Guide

github

: 17.7k

nvim-aider

Nvim-aider is a plugin for Neovim that provides additional functionality and key mappings to enhance the user's editing experience. It offers features such as code navigation, quick access to commonly used commands, and improved text manipulation tools. With Nvim-aider, users can streamline their workflow and increase productivity while working with Neovim.

github

: 86

llms

LLMs is a universal LLM API transformation server designed to standardize requests and responses between different LLM providers such as Anthropic, Gemini, and Deepseek. It uses a modular transformer system to handle provider-specific API formats, supporting real-time streaming responses and converting data into standardized formats. The server transforms requests and responses to and from unified formats, enabling seamless communication between various LLM providers.

github

: 168

fastapi_mcp

FastAPI-MCP is a zero-configuration tool that automatically exposes FastAPI endpoints as Model Context Protocol (MCP) tools. It allows for direct integration with FastAPI apps, automatic discovery and conversion of endpoints to MCP tools, preservation of request and response schemas, documentation preservation similar to Swagger, and the ability to extend with custom MCP tools. Users can easily add an MCP server to their FastAPI application and customize the server creation and configuration. The tool supports connecting to the MCP server using SSE or mcp-proxy stdio for different MCP clients. FastAPI-MCP is developed and maintained by Tadata Inc.

github

: 10.2k

PulsarRPAPro

PulsarRPAPro is a powerful robotic process automation (RPA) tool designed to automate repetitive tasks and streamline business processes. It offers a user-friendly interface for creating and managing automation workflows, allowing users to easily automate tasks without the need for extensive programming knowledge. With features such as task scheduling, data extraction, and integration with various applications, PulsarRPAPro helps organizations improve efficiency and productivity by reducing manual work and human errors. Whether you are a small business looking to automate simple tasks or a large enterprise seeking to optimize complex processes, PulsarRPAPro provides the flexibility and scalability to meet your automation needs.

github

: 102

context-portal

Context-portal is a versatile tool for managing and visualizing data in a collaborative environment. It provides a user-friendly interface for organizing and sharing information, making it easy for teams to work together on projects. With features such as customizable dashboards, real-time updates, and seamless integration with popular data sources, Context-portal streamlines the data management process and enhances productivity. Whether you are a data analyst, project manager, or team leader, Context-portal offers a comprehensive solution for optimizing workflows and driving better decision-making.

github

: 619

aide

Aide is a code-first API documentation and utility library for Rust, along with other related utility crates for web-servers. It provides tools for creating API documentation and handling JSON request validation. The repository contains multiple crates that offer drop-in replacements for existing libraries, ensuring compatibility with Aide. Contributions are welcome, and the code is dual licensed under MIT and Apache-2.0. If Aide does not meet your requirements, you can explore similar libraries like paperclip, utoipa, and okapi.

github

: 433

vector-cookbook

The Vector Cookbook is a collection of recipes and sample application starter kits for building AI applications with LLMs using PostgreSQL and Timescale Vector. Timescale Vector enhances PostgreSQL for AI applications by enabling the storage of vector, relational, and time-series data with faster search, higher recall, and more efficient time-based filtering. The repository includes resources, sample applications like TSV Time Machine, and guides for creating, storing, and querying OpenAI embeddings with PostgreSQL and pgvector. Users can learn about Timescale Vector, explore performance benchmarks, and access Python client libraries and tutorials.

github

: 111

deepflow

DeepFlow is an open-source project that provides deep observability for complex cloud-native and AI applications. It offers Zero Code data collection with eBPF for metrics, distributed tracing, request logs, and function profiling. DeepFlow is integrated with SmartEncoding to achieve Full Stack correlation and efficient access to all observability data. With DeepFlow, cloud-native and AI applications automatically gain deep observability, removing the burden of developers continually instrumenting code and providing monitoring and diagnostic capabilities covering everything from code to infrastructure for DevOps/SRE teams.

github

: 3.5k

authed

Authed is an identity and authentication system designed for AI agents, providing unique identities, secure agent-to-agent authentication, and dynamic access policies. It eliminates the need for static credentials and human intervention in authentication workflows. The protocol is developer-first, open-source, and scalable, enabling AI agents to interact securely across different ecosystems and organizations.

github

: 77

onlook

Onlook is a web scraping tool that allows users to extract data from websites easily and efficiently. It provides a user-friendly interface for creating web scraping scripts and supports various data formats for exporting the extracted data. With Onlook, users can automate the process of collecting information from multiple websites, saving time and effort. The tool is designed to be flexible and customizable, making it suitable for a wide range of web scraping tasks.

github

: 22.4k

AI-Codereview-Gitlab

AI-Codereview-Gitlab is an automated code review tool based on large models, designed to help development teams conduct intelligent code reviews quickly during code merging or submission. It supports multiple large models including DeepSeek, ZhipuAI, OpenAI, and Ollama. The tool can automatically push review results to DingTalk, WeChat Work, and Feishu, generate daily reports based on GitLab commit records, and provide a visual dashboard to display code review records. The tool works by triggering webhook events on GitLab when users submit code, calling third-party large models to review the code, and recording the review results in corresponding Merge Requests or Commit Notes.

github

: 735

MassGen

MassGen is a cutting-edge multi-agent system that leverages the power of collaborative AI to solve complex tasks. It assigns a task to multiple AI agents who work in parallel, observe each other's progress, and refine their approaches to converge on the best solution to deliver a comprehensive and high-quality result. The system operates through an architecture designed for seamless multi-agent collaboration, with key features including cross-model/agent synergy, parallel processing, intelligence sharing, consensus building, and live visualization. Users can install the system, configure API settings, and run MassGen for various tasks such as question answering, creative writing, research, development & coding tasks, and web automation & browser tasks. The roadmap includes plans for advanced agent collaboration, expanded model, tool & agent integration, improved performance & scalability, enhanced developer experience, and a web interface.

github

: 425

For similar tasks

mcp-documentation-server

github

: 201

Backlog.md

Backlog.md is a Markdown-native Task Manager & Kanban visualizer for any Git repository. It turns any folder with a Git repo into a self-contained project board powered by plain Markdown files and a zero-config CLI. Features include managing tasks as plain .md files, private & offline usage, instant terminal Kanban visualization, board export, modern web interface, AI-ready CLI, rich query commands, cross-platform support, and MIT-licensed open-source. Users can create tasks, view board, assign tasks to AI, manage documentation, make decisions, and configure settings easily.

github

: 3.2k

coco-app

Coco AI is a unified search platform that connects enterprise applications and data into a single, powerful search interface. The COCO App allows users to search and interact with their enterprise data across platforms. It also offers a Gen-AI Chat for Teams tailored to team's unique knowledge and internal resources, enhancing collaboration by making information instantly accessible and providing AI-driven insights based on enterprise's specific data.

github

: 281

For similar jobs

Azure-Analytics-and-AI-Engagement

The Azure-Analytics-and-AI-Engagement repository provides packaged Industry Scenario DREAM Demos with ARM templates (Containing a demo web application, Power BI reports, Synapse resources, AML Notebooks etc.) that can be deployed in a customer’s subscription using the CAPE tool within a matter of few hours. Partners can also deploy DREAM Demos in their own subscriptions using DPoC.

github

: 136

quivr

Quivr is a personal assistant powered by Generative AI, designed to be a second brain for users. It offers fast and efficient access to data, ensuring security and compatibility with various file formats. Quivr is open source and free to use, allowing users to share their brains publicly or keep them private. The marketplace feature enables users to share and utilize brains created by others, boosting productivity. Quivr's offline mode provides anytime, anywhere access to data. Key features include speed, security, OS compatibility, file compatibility, open source nature, public/private sharing options, a marketplace, and offline mode.

github

: 37.6k

Avalonia-Assistant

Avalonia-Assistant is an open-source desktop intelligent assistant that aims to provide a user-friendly interactive experience based on the Avalonia UI framework and the integration of Semantic Kernel with OpenAI or other large LLM models. By utilizing Avalonia-Assistant, you can perform various desktop operations through text or voice commands, enhancing your productivity and daily office experience.

github

: 113

MetaGPT

MetaGPT is a multi-agent framework that enables GPT to work in a software company, collaborating to tackle more complex tasks. It assigns different roles to GPTs to form a collaborative entity for complex tasks. MetaGPT takes a one-line requirement as input and outputs user stories, competitive analysis, requirements, data structures, APIs, documents, etc. Internally, MetaGPT includes product managers, architects, project managers, and engineers. It provides the entire process of a software company along with carefully orchestrated SOPs. MetaGPT's core philosophy is "Code = SOP(Team)", materializing SOP and applying it to teams composed of LLMs.

github

: 51.4k

UFO

UFO is a UI-focused dual-agent framework to fulfill user requests on Windows OS by seamlessly navigating and operating within individual or spanning multiple applications.

github

: 6.6k

timefold-solver

Timefold Solver is an optimization engine evolved from OptaPlanner. Developed by the original OptaPlanner team, our aim is to free the world of wasteful planning.

github

: 1.3k

MateCat

Matecat is an enterprise-level, web-based CAT tool designed to make post-editing and outsourcing easy and to provide a complete set of features to manage and monitor translation projects.

github

: 427

crewAI

crewAI is a cutting-edge framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks. It provides a flexible and structured approach to AI collaboration, enabling users to define agents with specific roles, goals, and tools, and assign them tasks within a customizable process. crewAI supports integration with various LLMs, including OpenAI, and offers features such as autonomous task delegation, flexible task management, and output parsing. It is open-source and welcomes contributions, with a focus on improving the library based on usage data collected through anonymous telemetry.

github

: 17.2k