What is Mcp O1js

mcp-o1js is a TypeScript package that serves as the core library for interacting with the Model Context Protocol (MCP) using O1JS. It is part of a monorepo that includes various packages for data ingestion and API gateway functionalities.

Use cases

Use cases for mcp-o1js include building applications that require real-time data ingestion from platforms like Discord, generating embeddings for machine learning models, and creating APIs for vector-based search functionalities.

How to use

To use mcp-o1js, you need to set up the project by running ‘npm install’ to install the necessary dependencies. After that, you can build all packages with ‘npm build’ and start developing by using the core library in your TypeScript applications.

Key features

Key features of mcp-o1js include data ingestion from multiple sources, integration with vector databases, and the ability to generate embeddings. It also provides a structured project setup with separate packages for different functionalities.

Where to use

mcp-o1js can be used in various fields such as data processing, machine learning, and application development where interaction with vector databases and data ingestion is required.

MCP o1js Project

This project provides TypeScript packages for o1js Model Context Protocol (MCP) server.

Project Structure

This is a monorepo containing multiple packages:

• packages/ingestion: A package for data ingestion from various sources
• packages/mcp-o1js: The core library for interacting with MCP using O1JS
• packages/vector-middleware: An API gateway for interacting with vector databases

Requirements

• Node.js
• Npm

Setup

To set up the project, run:

npm install

Build

To build all packages, run:

npm build

Development

Project Structure

packages/
├── ingestion/                    # Data ingestion package
│   └── src/
│       ├── index.ts              # Entry point
│       ├── config.ts             # Configuration
│       ├── ingestors/            # Data ingestors
│       │   ├── discord.ts        # Discord chat ingestion
│       │   ├── markdown.ts       # Markdown ingestion
│       │   └── typescript.ts     # TypeScript code ingestion
│       ├── embeddings/           # Embeddings generation
│       │   └── openai.ts         # OpenAI embedding service
│       └── storage/              # Vector storage
│           └── qdrant.ts         # Qdrant vector database adapter
├── mcp-server-o1js/              # MCP server package for o1js
│   └── src/
│       └── index.ts              # Entry point
└── vector-middleware/            # API Gateway for vector search
    └── src/
        ├── index.ts              # Entry point
        ├── qdrant.ts             # Qdrant integration
        ├── embedding.ts          # Embedding service
        ├── ratelimit.ts          # Rate limiting middleware
        └── config.ts             # Configuration