S3 Documentation MCP Server

S3 Documentation MCP Server is an MCP server that provides Retrieval-Augmented Generation capabilities over Markdown documentation stored in S3-compatible cloud storage. It connects client applications to buckets hosted on AWS S3, MinIO, Scaleway, DigitalOcean Spaces, or Cloudflare R2, enabling software developers, technical writers, and knowledge workers to query and retrieve technical documents directly from their development environments. The server parses remote Markdown files, chunks content, and generates vector embeddings using either local Ollama instances or OpenAI embedding models. It maintains an in-memory HNSWLib vector index stored as local files, tracking changes through ETag comparisons to synchronize updates incrementally. Through Model Context Protocol tools and native resources, AI assistants can perform semantic searches against documentation, fetch entire raw documents, and trigger indexing refreshes. This setup allows engineering teams to keep language models grounded in their most up-to-date internal wikis, product manuals, and API references without operating complex vector database clusters.

Category: AI Memory & Context

Tags: aws, documentation, markdown, rag, s3

Visit S3 Documentation MCP Server

How to install and configure S3 Documentation MCP Server

1. Clone the repository and configure your environment: bash cp env.example .env Edit .env to set your S3 credentials (S3_BUCKET_NAME, S3_ACCESS_KEY_ID, S3_SECRET_ACCESS_KEY, S3_REGION) and choose your embedding provider (ollama or openai). 2. Run the server using Docker: bash docker run -d \ --name s3-doc-mcp \ -p 3000:3000 \ --env-file .env \ -e OLLAMA_BASE_URL=http://host.docker.internal:11434 \ -v $(pwd)/data:/app/data \ yoanbernabeu/s3-doc-mcp:latest Alternatively, install dependencies locally with npm install and launch using npm run build && npm start. 3. Register the server in Cursor (~/.cursor/mcp.json) or Claude Desktop (claude_desktop_config.json): json { "mcpServers": { "doc": { "type": "streamable-http", "url": "http://127.0.0.1:3000/mcp", "note": "S3 Documentation RAG Server" } } } Restart your MCP client to access the tools.

What you can do with S3 Documentation MCP Server

• Search internal team wikis stored in MinIO or S3 directly through Claude Desktop without manually uploading files. - Query up-to-date API references and SDK guides within Cursor to assist real-time coding and implementation. - Enable AI coding assistants to retrieve full system architectural documents using Markdown chunking and semantic vector search. - Ground developer responses using local embeddings via Ollama to ensure complete data privacy during documentation lookup.

Key facts

• https://github.com/yoanbernabeu/S3-Documentation-MCP-Server
• AI Memory & Context, Files, Documents & PDFs
• aws, documentation, markdown, rag, s3

Part of MCP Servers

What can S3 Documentation MCP Server do?

The server enables AI assistants to perform semantic vector searches across Markdown documentation hosted on S3-compatible storage. It exposes three primary MCP tools: searching indexed chunks via cosine similarity, retrieving entire raw documents, and manually refreshing the index. It also exposes standard MCP Resources for file discovery.

Which MCP clients work with S3 Documentation MCP Server?

The server exposes a streamable HTTP endpoint that integrates with clients supporting remote Model Context Protocol connections, including Claude Desktop and Cursor. Users add the server URL to their client configuration files to register tools and resources.

Which embedding providers are supported?

It supports both local and cloud-based embedding providers. You can run locally with Ollama using the nomic-embed-text model for private and offline operation, or connect to the OpenAI API using models like text-embedding-3-small and text-embedding-3-large.

What storage providers can I use with this server?

It is compatible with standard AWS S3 as well as alternative S3-compatible object storage providers including MinIO, Scaleway, DigitalOcean Spaces, Wasabi, and Cloudflare R2 by configuring the custom S3 endpoint variable.