What is RAGStack and how does this MCP server connect to it?

RAGStack is a serverless knowledge base platform on AWS. This MCP server acts as a bridge, letting you search, upload documents, and chat with your RAGStack instance via Claude or other AI tools. You connect it by providing your GraphQL endpoint and API key from the RAGStack Dashboard.

How do I get my RAGSTACK_GRAPHQL_ENDPOINT and RAGSTACK_API_KEY?

Both credentials are available in your RAGStack Dashboard under Settings. Copy the GraphQL API URL and generate or retrieve your API key, then set them as environment variables when running the MCP server.

Can I use this MCP server to upload and search documents in the same conversation?

Yes. The server supports upload, search, and chat capabilities, so you can upload a document and immediately search or query it within the same session.

What transport protocol does this MCP server use?

It uses stdio (standard input/output), which is the most common transport for local MCP integrations. Install it via `uvx ragstack-mcp@0.1.4` and configure your AI tool to spawn it as a subprocess.

Is there a cost to using RAGStack with this MCP server?

RAGStack is a managed AWS service with its own pricing model. Check the RAGStack documentation or AWS console for current pricing; the MCP server itself is open-source and free.

MCP ServerSTDIOOfficialv0.1.4

Hatmanstack Ragstack MCP Server

Search, chat, and upload documents to a serverless RAGStack knowledge base on AWS. For developers building AI applications that need semantic search and retrieval-augmented generation without managing infrastructure.

io.github.HatmanStack/ragstack

Try MCP with AI models View source

Hosted URL

Local install

Transport

STDIO

Auth

No auth required

What the Hatmanstack Ragstack MCP server does

How models use it and what it is built for.

Connect to Hatmanstack Ragstack

Local install — runs as a subprocess.

uvx ragstack-mcp==0.1.4

Environment variables

Configuration this server reads at startup.

RAGSTACK_GRAPHQL_ENDPOINT
Your RAGStack GraphQL API URL (from Dashboard → Settings)
RAGSTACK_API_KEY
Your RAGStack API key (from Dashboard → Settings)

Resources

Where to find authoritative docs and source for Hatmanstack Ragstack.

Source repository

Example prompts for Hatmanstack Ragstack

Paste any of these into Agent Studio after connecting Hatmanstack Ragstack.

Search the RAGStack knowledge base for documents about authentication patterns
Upload a PDF file to the RAGStack knowledge base and index it
Chat with the RAGStack knowledge base to answer questions about our API
What GraphQL endpoint and API key format does RAGStack expect in the MCP configuration

Hatmanstack Ragstack MCP server — FAQ

Common questions about connecting and running Hatmanstack Ragstack.

What is RAGStack and how does this MCP server connect to it?
RAGStack is a serverless knowledge base platform on AWS. This MCP server acts as a bridge, letting you search, upload documents, and chat with your RAGStack instance via Claude or other AI tools. You connect it by providing your GraphQL endpoint and API key from the RAGStack Dashboard.
How do I get my RAGSTACK_GRAPHQL_ENDPOINT and RAGSTACK_API_KEY?
Both credentials are available in your RAGStack Dashboard under Settings. Copy the GraphQL API URL and generate or retrieve your API key, then set them as environment variables when running the MCP server.
Can I use this MCP server to upload and search documents in the same conversation?
Yes. The server supports upload, search, and chat capabilities, so you can upload a document and immediately search or query it within the same session.
What transport protocol does this MCP server use?
It uses stdio (standard input/output), which is the most common transport for local MCP integrations. Install it via `uvx ragstack-mcp@0.1.4` and configure your AI tool to spawn it as a subprocess.
Is there a cost to using RAGStack with this MCP server?
RAGStack is a managed AWS service with its own pricing model. Check the RAGStack documentation or AWS console for current pricing; the MCP server itself is open-source and free.

Skip the local setup — run MCP in your browser

MCP Playground runs 10,000+ hosted MCP servers — GitHub, Linear, Notion, Stripe, Sentry and more — across Claude, GPT, Gemini, DeepSeek and 60+ AI models. Compare model answers side-by-side, save agent presets, share runs. Zero install.

Open Agent Studio