christopher-buss/roblox-docs-mcp
If you are the rightful owner of roblox-docs-mcp and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
This MCP server provides intelligent search and retrieval capabilities for Roblox documentation using semantic search and vector embeddings.
๐ Roblox Documentation MCP Server with RAG Support
MCP server with RAG support for intelligent Roblox documentation search and retrieval
This MCP server enables AI agents to intelligently search and retrieve Roblox documentation through semantic search and vector embeddings. It provides natural language access to the complete Roblox Creator Documentation.
๐ฏ What This Does
Enable AI agents to:
- ๐ Semantic Search: Find relevant documentation through natural language queries
- ๐ API References: Get specific details about Roblox classes, methods, and properties
- ๐ Tutorial Discovery: Locate step-by-step guides and learning materials
- ๐ก Code Examples: Find relevant code snippets and demonstrations
- ๐ท๏ธ Smart Filtering: Search by content type, difficulty, or topic
๐๏ธ Architecture
graph TD
A[AI Agent] --> B[MCP Server]
B --> C[RAG Service]
C --> D[ChromaDB Vector Store]
C --> E[OpenAI Embeddings]
B --> F[Git Service]
F --> G[Roblox/creator-docs Repository]
B --> H[Content Processor]
H --> I[Markdown Parser]
H --> J[YAML Parser]
โจ Key Features
Feature Area | Description | Implementation |
---|---|---|
๐ Semantic Search | Natural language queries across all Roblox documentation | ChromaDB + OpenAI embeddings |
๐ Content Processing | Processes markdown guides, tutorials, and YAML API references | markdown-it + yaml parsers |
๐ Auto-Updates | Keeps documentation current via git pull from official repository | simple-git integration |
๐ท๏ธ Smart Classification | Automatically categorizes content (guides, tutorials, API references) | Metadata extraction + classification |
โก Performance | Fast semantic search with caching and optimized vector storage | Redis caching + ChromaDB |
๐ Production Ready | Built on proven MCP template with comprehensive error handling | Full TypeScript + Zod validation |
๐ Quick Start
Prerequisites
- Node.js 20+
- ChromaDB server (Docker recommended)
- OpenAI API key
Installation
# Clone the repository
git clone https://github.com/christopher-buss/roblox-docs-mcp.git
cd roblox-docs-mcp
# Install dependencies
npm install
# Set up environment variables
cp .env.example .env
# Edit .env with your OpenAI API key and ChromaDB settings
# Start ChromaDB (using Docker)
docker run -p 8000:8000 chromadb/chroma
# Build the project
npm run build
Environment Configuration
Create a .env
file with the following variables:
# OpenAI Configuration
OPENAI_API_KEY=your_openai_api_key
# ChromaDB Configuration
CHROMA_DB_URL=http://localhost:8000
CHROMA_DB_COLLECTION=roblox-docs
# Roblox Documentation
ROBLOX_DOCS_REPO_URL=https://github.com/Roblox/creator-docs.git
ROBLOX_DOCS_LOCAL_PATH=./data/roblox-docs
ROBLOX_DOCS_UPDATE_INTERVAL=24
# Embedding Configuration
EMBEDDING_MODEL=text-embedding-3-large
MAX_CHUNK_SIZE=1000
CHUNK_OVERLAP=200
# Optional Redis Cache
REDIS_URL=redis://localhost:6379
Running the Server
# Start MCP server (stdio transport)
npm run start:stdio
# Start MCP server (HTTP transport)
npm run start:http
# Update documentation repository
npm run docs:update
# Launch MCP inspector for debugging
npm run inspector
๐ ๏ธ Available MCP Tools
searchRobloxDocs
Purpose: Semantic search across all Roblox documentation
Input: Natural language query, optional filters
Output: Ranked list of relevant documentation with metadata
getRobloxApiReference
Purpose: Get specific API class/method documentation
Input: API name, class name, method name
Output: Detailed API documentation with examples
findRobloxTutorials
Purpose: Find step-by-step tutorials and guides
Input: Topic, difficulty level, tutorial type
Output: Curated list of tutorials with descriptions
getRobloxGuides
Purpose: Retrieve conceptual guides and explanations
Input: Topic area, content type
Output: Relevant guides with structured content
๐ Project Structure
src/
โโโ services/
โ โโโ git-service/ # Git repository operations
โ โโโ content-processor/ # Markdown/YAML processing
โ โโโ roblox-rag/ # RAG implementation
โโโ mcp-server/
โ โโโ tools/ # MCP tools for documentation search
โ โโโ server.ts # Main MCP server
โโโ config/ # Configuration management
โโโ utils/ # Production utilities
๐ง Development
Architecture Overview
This project extends the cyanheads/mcp-ts-template
with Roblox-specific capabilities:
- Git Service: Manages the Roblox creator-docs repository
- Content Processor: Parses markdown and YAML files
- RAG Service: Handles embeddings and semantic search
- MCP Tools: Provides search and retrieval capabilities
Adding New Features
- New Tools: Follow the template pattern in
src/mcp-server/tools/
- Content Processing: Extend processors in
src/services/content-processor/
- RAG Enhancements: Modify search logic in
src/services/roblox-rag/
Development Commands
npm run build # Build TypeScript
npm run format # Format code with Prettier
npm run docs:generate # Generate TypeDoc documentation
npm run tree # Generate project structure
npm run depcheck # Check for unused dependencies
๐งช Testing
# Test individual components
npm run test:unit
# Test MCP tools end-to-end
npm run test:integration
# Test RAG functionality
npm run test:rag
๐ Performance
- Search Latency: < 500ms for semantic queries
- Memory Usage: < 2GB RAM for full documentation index
- Document Processing: 100+ docs/minute ingestion rate
- Cache Hit Rate: > 80% for repeated queries
๐ค Contributing
- Fork the repository
- Create a feature branch
- Make your changes
- Add tests if applicable
- Submit a pull request
๐ License
This project is licensed under the Apache License 2.0. See the file for details.
๐ Acknowledgments
- Built on the excellent cyanheads/mcp-ts-template
- Powered by Roblox Creator Documentation
- Uses ChromaDB for vector storage
- Embeddings by OpenAI
๐ Documentation
- - Complete implementation roadmap
- - Guide for AI development
- - Claude Code specific guidance
Note: This project is currently in development. See for current status and implementation progress.