MCP Integration

Your Corpus in Every Tool

CaveauAI implements the Model Context Protocol. Query your private documents from Claude Desktop, Cursor, VS Code, or any MCP client.

What is MCP?

The Model Context Protocol (MCP) is an open standard that allows AI assistants to connect to external data sources and tools. Think of it as a USB port for AI — a universal interface that lets any compatible AI assistant access your CaveauAI corpora.

When you connect CaveauAI as an MCP server, your AI assistant gains the ability to search your private documents, retrieve cited answers, and even upload new documents — all from within your existing workflow.

Available MCP Tools

corpus_search

Search across one or more corpora using natural language. Returns relevant passages with source citations.

corpus_chat

Ask a question and get a synthesized answer grounded in your documents, with full citation chains.

document_upload

Upload a new document to a corpus. The pipeline processes it automatically.

corpus_list

List all corpora available to your account, with document counts and status.

Configuration

Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "bluenote-ai": {
      "url": "https://ai.bluenotelogic.com/mcp",
      "headers": {
        "Authorization": "Bearer bnai_sk_your_key_here"
      }
    }
  }
}

Cursor / VS Code

Add to your project's .mcp.json:

{
  "servers": {
    "bluenote-ai": {
      "type": "sse",
      "url": "https://ai.bluenotelogic.com/mcp/sse",
      "headers": {
        "Authorization": "Bearer bnai_sk_your_key_here"
      }
    }
  }
}

Security

  • All MCP connections use TLS 1.3 encryption
  • API keys are scoped to specific corpora — you control access
  • No document content is cached by the MCP layer
  • All queries are logged to your audit trail

Need help with MCP integration?