Procedural Memory Server

This server is the agent's central memory, providing a searchable directory of all available tools and procedures to find the right resource for any task.

Features

• Semantic search across a knowledge base of documents
• Retrieval of full document content by ID
• Typesense integration for powerful and fast search

Prerequisites

• Node.js (v18 or higher)
• Typesense

Configuration

The server is configured using a .env file in the root of the project. You can use the .env.example file as a template.

# Typesense configuration
TYPESENSE_HOST="your-typesense-host"
TYPESENSE_PORT="8108"
TYPESENSE_PROTOCOL="http"
TYPESENSE_API_KEY="your-typesense-api-key"

Installation

1. Clone the repository:

   git clone https://github.com/syia-ai/mcp_procedural_memory.git
   cd mcp_procedural_memory

2. Install dependencies:

   npm install

3. Build the project:

   npm run build

4. Start the server:

   npm start

Development

For development with hot reload:

npm run dev

Testing

Test the server with the MCP Inspector:

npm test

You can also run the full suite of tests defined in test/test-config.yaml with the following command:

npm run test:mcp

MCP Tools

Document Tools

• find_relevant_documents: Performs a broad semantic search across the knowledge base to find the most relevant documents.
• get_full_document_by_id: Retrieves the complete, original content of a single document using its unique ID.

Project Structure

src/
├── tools/
│   ├── handlers/
│   │   └── tools.ts    # Handlers for the available tools
│   ├── index.ts        # Main tool handler
│   └── schema.ts       # Tool schemas
├── types/
│   └── index.ts        # TypeScript type definitions
└── index.ts            # Main application entry point

Contributing

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add some amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.