awslabs.aws-documentation-mcp-server

What it is and what it does

This package is an MCP (Model Context Protocol) server that bridges AWS documentation and AI assistants. It runs as a subprocess in compatible editors and IDEs (Cursor, VS Code, Kiro) and exposes AWS documentation as a set of tools: read full pages as markdown, search documentation using AWS's official search API, fetch specific sections, filter large tables without loading entire pages, and get content recommendations. It supports both global AWS and AWS China documentation partitions via environment configuration.

The server is built on standard web-fetching libraries (httpx, beautifulsoup4, markdownify) to scrape and convert AWS documentation to markdown, and uses pydantic for structured responses. It's designed for developers who want their AI coding assistant to have real-time access to AWS documentation without manual lookups. Installation is straightforward via pip or uv, and configuration is a JSON block in your client's MCP settings.

Use it for:

• Ask an AI assistant to look up AWS S3 bucket naming rules or service quotas without leaving your editor.
• Search AWS documentation for a specific feature or service within an IDE conversation without context switching.
• Retrieve and filter large AWS pricing or service-quota tables by specific criteria without loading the full page.
• Get content recommendations for an AWS documentation page to discover related guides and best practices.
• Query AWS China documentation partition when working with AWS GovCloud or China region deployments.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Provides an MCP server that exposes AWS documentation as searchable tools for AI assistants and code editors, enabling retrieval, search, and recommendations across AWS service guides.

Yes, if you use an MCP-compatible editor (Cursor, VS Code, Kiro) and want your AI assistant to access live AWS documentation. Installation is low-friction, the package is actively maintained, and the Apache-2.0 license poses no restrictions. The only gotcha is that it requires Python 3.10+ and runs as a server subprocess, not a library you import directly—so it's only useful within an MCP client, not in standalone scripts.

Install

awslabs-aws-documentation-mcp-server on PyPI

pip install awslabs-aws-documentation-mcp-server

uv add awslabs-aws-documentation-mcp-server

poetry add awslabs-aws-documentation-mcp-server

Installing awslabs.aws-documentation-mcp-server

Before you install

Low friction install via wheel distribution. Active maintenance with recent release (4 days old) and high repository engagement (9598 stars). Requires Python 3.10 or newer and depends on six common libraries (httpx, beautifulsoup4, pydantic, loguru, markdownify, mcp).

License in practice

Apache-2.0 permissive license allows commercial use, modification, and distribution with minimal restrictions—suitable for most projects.

Quickstart

# Install via uv or pip
pip install awslabs.aws-documentation-mcp-server

# Configure in your MCP client (e.g., Cursor, VS Code, Kiro):
# Add to mcpServers config with command: uvx awslabs.aws-documentation-mcp-server@latest
# Set env vars: FASTMCP_LOG_LEVEL=ERROR, AWS_DOCUMENTATION_PARTITION=aws

Requires Python 3.10 or newer. Designed to run as an MCP server subprocess (not a direct library import); intended for use within MCP-compatible clients like Cursor, VS Code, or Kiro, not standalone Python scripts.

Verify before relying

• Whether search and recommendation features work reliably across all AWS service categories and documentation page types.
• Performance characteristics when searching large documentation tables or pages with hundreds of rows.
• Behavior and fallback handling when behind corporate proxies or firewalls that block User-Agent strings.

Package facts

Evidence: awslabs_aws_documentation_mcp_server-1.1.30-py3-none-any.whl

Tags

More Documentation packages

Further reading