dryad-rag-mcp
OfficialClick on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@dryad-rag-mcp Find how to increase the top_k limit in the docs"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
dryad-rag-mcp
A small Model Context Protocol server that exposes a
RAG document-search API as agent-callable tools — rag_health, rag_ingest, and
rag_query. It wraps dryad-rag-pipeline,
so any MCP client (Claude Code, Claude Desktop, or your own client) can index and search
markdown docs through tool calls instead of hand-rolled HTTP glue.
Why this exists
Most MCP server examples are toys that echo a string back. This one wraps a real, separately running HTTP service and is proven against it: the test suite spins up the actual FastAPI RAG service, spawns this server as a subprocess, connects a real MCP client over stdio, and asserts the retrieved chunks are semantically correct — not just that the tool call didn't crash.
Related MCP server: mcp-agentic-rag
Tools
Tool | Description |
| Checks whether the RAG service is reachable. |
| Re-indexes the configured document set. |
| Searches the index; returns cited chunks with a similarity score. |
Quickstart
npm install
npm run buildPoint it at a running dryad-rag-pipeline instance:
RAG_API_BASE_URL=http://127.0.0.1:8000 npm startAdd to Claude Code
claude mcp add dryad-rag -- node /path/to/dryad-rag-mcp/dist/server.jsOr in .mcp.json:
{
"mcpServers": {
"dryad-rag": {
"command": "node",
"args": ["/path/to/dryad-rag-mcp/dist/server.js"],
"env": { "RAG_API_BASE_URL": "http://127.0.0.1:8000" }
}
}
}Running the tests
The test suite needs a checkout of dryad-rag-pipeline as a sibling directory (or set
RAG_PIPELINE_DIR), with its Python virtualenv already set up:
git clone https://github.com/DryadAI/dryad-rag-pipeline ../dryad-rag-pipeline
cd ../dryad-rag-pipeline && python3 -m venv .venv && .venv/bin/pip install -r requirements-dev.txt
cd ../dryad-rag-mcp
npm run build
npm testThis starts the real RAG API on a test port, spawns this MCP server as a subprocess,
connects an MCP Client over stdio, and exercises all three tools end to end — including
asserting that a health question actually retrieves the FAQ/API-reference chunks that
answer it, and that an out-of-range top_k is rejected.
Architecture
sequenceDiagram
participant Agent as MCP Client (e.g. Claude)
participant Server as dryad-rag-mcp (stdio)
participant API as dryad-rag-pipeline (HTTP)
Agent->>Server: callTool("rag_ingest")
Server->>API: POST /ingest
API-->>Server: {documents_ingested, chunks_created}
Server-->>Agent: "Indexed 3 document(s)..."
Agent->>Server: callTool("rag_query", {question})
Server->>API: POST /query
API-->>Server: ranked, cited chunks
Server-->>Agent: formatted, citable textLicense
MIT — see LICENSE.
Available Tools
3 toolsrag_healthRAG service health checkA
Check whether the dryad-rag-pipeline service at https://dryad-rag-pipeline.example.com is reachable.
Inputs: no named parameters.
rag_ingestIngest documents into the RAG indexA
Re-index the configured markdown document set on the RAG service. Call this once before rag_query if the index is empty or the docs changed.
Inputs: no named parameters.
rag_queryQuery the RAG indexA
Ask a question against the indexed documents and get back the most relevant chunks with their source file and a similarity score, for citation.
Inputs: question (required), top_k.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 3 tools
Each tool has a clearly distinct purpose: health check, ingest/re-index, and query. There is no overlap or ambiguity between them.
All tools follow a consistent 'rag_' prefix with a verb (health, ingest, query), forming a uniform and predictable naming pattern.
Three tools is well-scoped for a focused RAG service, covering the core operations of health checking, indexing, and querying without excess.
The set covers the essential lifecycle for a RAG service: health, ingest, and query. A minor gap is lack of explicit index management (e.g., clear or list documents), but the configured document set makes ingest sufficient for updates.
Related MCP Connectors
Ingest, manage, and retrieve documents for RAG-powered AI applications
Cloud or self-hosted knowledge for AI agents: hybrid search, reranking, GraphRAG, scoped MCP tools.
Team docs served to AI agents over MCP - search, Markdown reads, version pinning, read audit.
Self-hosted AI workspace: semantic RAG document search, long-term memory, and AI chat via MCP.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceExposes document retrieval as an MCP tool, enabling LLMs to search a local vector store of markdown documents. Includes a retrieval evaluation harness to measure hit rate and MRR.MIT
- FlicenseCqualityDmaintenanceProvides RAG tools with local vector retrieval and web fallback using Firecrawl, enabling document ingestion and querying through MCP stdio transport.3-
- FlicenseNot gradedqualityDmaintenanceExposes an existing RAG API as MCP tools, enabling health checks and document-based question answering with source evidence.-
- AlicenseNot gradedqualityDmaintenanceEnables retrieval-augmented generation over a local markdown corpus, allowing grounded, cited answers via an MCP tool or CLI.MIT