Vercel Ai Docs MCP

🚀 Vercel AI SDK Documentation MCP Agent

A toolset for querying and synthesizing information from the Vercel AI SDK documentation.

🚀 Quick Start

The Vercel AI SDK Documentation MCP Agent is a toolset designed to query and synthesize information from the Vercel AI SDK documentation. Here is the complete project documentation.

✨ Features

This project provides a server based on the Model Context Protocol (MCP), supporting queries to the Vercel AI SDK documentation through intelligent agents and direct searches. The main functions include:

1. agent-query: Use an AI agent to retrieve and summarize documentation information.
2. direct-query: Conduct a similarity search directly on the documentation content.
3. clear-memory: Clear the context memory of a specific session or all sessions.

Key Features

• Support for multi-round conversation context, facilitating the resolution of complex problems.
• Flexible query parameters (e.g., limit the number of results) to meet different needs.
• Built-in document indexing function, supporting fast retrieval and loading.
• Support for session management, ensuring that each interaction is independent and traceable.

🔧 Technical Details

Architecture Components

├── config/            # Configuration folder
├── core/              # Core functional module
│   ├── indexing/      # Document indexing and vector storage
│   └── query/         # Query services (agent and direct)
├── files/             # Storage directory
│   ├── docs/          # Processed documents
│   ├── faiss_index/   # Vector index files
│   └── sessions/      # Session data
├── mcp/               # MCP server and tools
│   ├── server.ts       # MCP server implementation
│   └── tools/          # MCP tool definitions
├── scripts/           # Build and utility scripts
└── utils/             # Auxiliary utility functions

📦 Installation

Environment Requirements

• Node.js (version 14 or higher)
• TypeScript (version 4.0 or higher)

Installation Steps

1. Clone the project repository:

git clone [REPOSITORY_URL]
cd Vercel-AI-SDK-Documentation-MCP-Agent

2. Install dependencies:

npm install

3. Compile the code:

npm run build

4. Build the document index (required for first-time use):

npm run build:index

📚 Documentation

Integration Guide

Integration with Claude Desktop

1. Create or edit the mcp-config.json file in the ~/.config/claude directory and add the following content:

{
  "tools": [
    {
      "name": "Vercel-AI-SDK-Documentation-MCP-Agent",
      "path": "/absolute/path/to/server"
    }
  ]
}

2. Start Claude Desktop and verify that the tool is working.

Integration with Cursor

1. Create or edit the mcp-config.json file in the project root directory and add the following content:

{
  "tools": [
    {
      "name": "Vercel-AI-SDK-Documentation-MCP-Agent",
      "path": "/absolute/path/to/server"
    }
  ]
}

2. Restart the Cursor application to load the configuration.

💻 Usage Examples

Basic Usage

1. agent-query

Query documentation information through an AI agent:

{
  "name": "agent-query",
  "arguments": {
    "query": "How to use the streamText function?",
    "sessionId": "Unique session identifier"
  }
}

2. direct-query

Conduct a similarity search directly on the documentation content:

{
  "name": "direct-query",
  "arguments": {
    "query": "Usage of streamText",
    "limit": 5
  }
}

3. clear-memory

Clear the context memory of a specific session or all sessions:

{
  "name": "clear-memory",