CB Insights MCP Server

What is CB Insights MCP Server about?

The server exposes a ChatCBI tool that lets agents send messages to the CB Insights LLM and receive structured responses, including chat IDs, related content, sources, and suggested prompts.

How to use CB Insights MCP Server?

1. Install dependencies using the Python package manager uv (as described in the README).
2. Copy the .env template and set the required environment variables:
   • CBI_CLIENT_ID and CBI_CLIENT_SECRET for API authentication.
   • CBI_MCP_PORT (default 8000) to change the listening port.
   • CBI_MCP_TIMEOUT to adjust request timeout.
3. Run the server, e.g., uv run server.py.
4. For Claude Desktop integration, execute mcp install server.py; this writes a claude_desktop_config.json entry that points to the uv command.
5. Debug with the Model Context Protocol inspector: mcp dev server.py.

Key Features

• ChatCBI tool: send a message and optionally a chatID; receive response details, related content, sources, and suggestions.
• Configurable port and timeout via .env.
• Built‑in authentication using CB Insights client credentials.
• Seamless Claude Desktop integration through generated JSON configuration.
• Inspector support for rapid development and testing.

Use Cases

• Building custom AI agents that need to query CB Insights market intelligence.
• Automating research workflows where continuity of conversation (chatID) is required.
• Creating chat‑driven dashboards or internal tools that surface data‑backed insights.
• Extending Claude Desktop with domain‑specific knowledge from CB Insights.

FAQ

Q: Which Python version is needed? A: Any version supported by uv; typically Python 3.9+.

Q: How do I obtain the client ID and secret? A: Register an application on the CB Insights developer portal and follow the authentication docs linked in the README.

Q: Can I change the server’s host address? A: Yes, set the HOST variable in the .env file (defaults to 0.0.0.0).

Q: What happens if I don’t provide a chatID? A: The server creates a new ChatCBI session and returns a fresh chatID in the response.

Q: Is there a way to view logs? A: Run the server with the --log-level DEBUG flag (or set LOG_LEVEL in .env).

CB Insights MCP Server

The CBI MCP Server provides an interface for developers to interact with CB Insights ChatCBI LLM through AI Agents.

Tools

ChatCBI

Send a message from an agent to ChatCBI and return the response.

• Input parameters:
  • message: The content of your message to ChatCBI
  • chatID(optional): A unique identifier for the chat session, obtained from a previous response. If included, the conversation is continued. Otherwise, a new conversation is started.
• Returns JSON object containing the following fields:
  • chatID: Identifies the conversation. If chatID was provided in the request, this will be the same.
  • message: ChatCBI response to the message.
  • relatedContent: List of related references.
  • sources: List of sources used to generate the response.
  • suggestions: List of suggested follow-up questions.
  • title: Title of the chat
• For more information see Using ChatCBI

Setup

The CBI MCP Server uses uv to manage the project.

Environment variables are set via the .env file:

• CBI_CLIENT_ID & CBI_CLIENT_SECRET OAuth Client ID and Secret
  • see CB Insights API Authentication for usage and how to obtain them
• CBI_MCP_TIMEOUT (default: 120)
• CBI_MCP_PORT (default: 8000)

Usage

With Claude Desktop

Update the claude_desktop_config.json file using the following command:

mcp install server.py

This will add the following configuration:

{
  "mcpServers": {
    "cbi-mcp-server": {
      "command": "/path/to/.local/bin/uv",
      "args": [
        "--directory",
        "/path/to/cloned/cbi-mcp-server",
        "run",
        "server.py"
      ]
    }
  }
}

Debugging

The inspector can be used to test/debug your server.

mcp dev server.py 

More info on using the inspector