Godot Documentation RAG

Provides an interactive, retrieval-augmented Q&A interface over Godot official docs using a local vector store and DeepSeek-backed answers.

python

5

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "lvsedeqinglvse--fastmcp-godot-rag-": {
      "command": "python",
      "args": [
        "rag_fastmcp.py"
      ],
      "env": {
        "DEEPSEEK_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

This MCP server lets you interactively query and interpret Godot engine documentation by combining a retrieval-augmented generation (RAG) workflow with a local knowledge base. You install it, provide your API key, and then ask questions as if you’re chatting with a knowledgeable helper who can surface precise Godot docs and guidance.

How to use

You run the local MCP server and connect your MCP client to it. The server exposes two main tools: search_godot_docs, which retrieves relevant document fragments, and rag_answer, which composes structured answers using the retrieved content and the DeepSeek API. Start the server with your API key set, then issue questions to obtain structured, source-informed answers about Godot documentation.

How to install

# Prerequisites
# Ensure you have Python installed on your system.
# You will also need to provide your DeepSeek API key when running the server.

# Step 1: Prepare to run the server
# Place any local data/assets needed by the server in your working directory as required by the project.

# Step 2: Start the MCP server
# The server is started by executing the tooling script that wires the local docs into the MCP interface.
# Set your API key as an environment variable before starting.
export DEEPSEEK_API_KEY=YOUR_API_KEY

# Step 3: Run the server entry point
python rag_fastmcp.py

Additional sections

Configuration: The server uses a local vector index and a metadata store to power fast retrieval and accurate answers. Ensure your local assets (documentation chunks, index, and metadata) are prepared as described by the project tooling, and provide your API key before starting.

Security and secrets: Treat your DeepSeek API key as sensitive. Do not commit it to version control. Set it in your environment before launching the server.

Examples of use: After starting, you can ask questions like “Show me how to implement a 2D physics scene in Godot,” or “What is the recommended way to structure a Godot project for 4.x?” The server uses the built-in search and answer tools to present concise, cited responses.

Available tools

search_godot_docs

Retrieves relevant Godot documentation fragments from the local vector store to support accurate, context-rich answers.

rag_answer

Generates structured, citation-backed answers by combining retrieved fragments with the DeepSeek API.