Back to MCP Catalog

Comet Opik MCP Server

Developer ToolsTypeScript
Access and manage Opik platform capabilities through a unified interface
Available Tools

prompts.list

List all prompts in the workspace

prompts.create

Create a new prompt in the workspace

prompts.update

Update an existing prompt

prompts.delete

Delete a prompt from the workspace

projects.list

List all projects/workspaces

traces.track

Track and analyze trace data

metrics.query

Gather and query metrics data

Comet Opik MCP provides a standardized interface to the Opik platform, enabling seamless integration with IDEs and development environments. It offers comprehensive management of prompts, projects, traces, and metrics for LLM applications through the Model Context Protocol (MCP). With support for multiple transport mechanisms including stdio and SSE, Opik MCP allows developers to access Opik's capabilities directly from their preferred development environment. This integration streamlines the workflow for organizing, monitoring, and optimizing LLM applications.

Overview

Comet Opik MCP Server provides a Model Context Protocol implementation for the Opik platform, allowing you to access Opik's capabilities directly from compatible IDEs and applications. This integration enables efficient management of prompts, projects, traces, and metrics for your LLM applications.

Installation

There are two main ways to install and use the Opik MCP Server:

Method 1: Manual Installation

  1. Clone the repository:

    git clone https://github.com/comet-ml/opik-mcp.git
    cd opik-mcp
    
  2. Install dependencies and build:

    npm install
    npm run build
    
  3. Create a configuration file:

    cp .env.example .env
    
  4. Edit the .env file with your specific configuration, including your Opik API key.

Method 2: Direct Integration with IDE

For direct integration with compatible IDEs like Cursor, you can configure the MCP server to run directly without manual installation.

Configuration

The Opik MCP Server requires the following configuration parameters:

  • OPIK_API_BASE_URL: The base URL for the Opik API (default: https://www.comet.com/opik/api)
  • OPIK_API_KEY: Your Opik API key
  • OPIK_WORKSPACE_NAME: The workspace name to use (default: "default")
  • DEBUG: Set to "true" to enable debug logging (optional)

Usage

Starting the Server

You can start the server using one of the following transport mechanisms:

Standard Input/Output (stdio)

This is the recommended approach for local integration:

npm run start:stdio

Or using the Makefile:

make start-stdio

Server-Sent Events (SSE) - Experimental

For remote access and multiple simultaneous clients:

npm run start:sse

Or using the Makefile:

make start-sse

IDE Integration

Cursor Integration

To integrate with Cursor IDE, create a .cursor/mcp.json file in your project directory with the following configuration:

{
  "mcpServers": {
    "opik": {
      "command": "/path/to/node",
      "args": [
        "/path/to/opik-mcp/build/index.js",
        "--apiUrl",
        "https://www.comet.com/opik/api",
        "--apiKey",
        "YOUR_API_KEY",
        "--workspace",
        "default",
        "--debug",
        "true"
      ],
      "env": {
        "OPIK_API_BASE_URL": "https://www.comet.com/opik/api",
        "OPIK_API_KEY": "YOUR_API_KEY",
        "OPIK_WORKSPACE_NAME": "default"
      }
    }
  }
}

Replace the following:

  • /path/to/node with the path to your Node.js executable
  • /path/to/opik-mcp with the path to your opik-mcp installation
  • YOUR_API_KEY with your actual Opik API key

Advanced Configuration

For advanced configuration options, refer to the documentation in the repository:

  • Transport-specific configuration
  • API endpoint customization
  • Authentication options
  • Logging and debugging settings

Troubleshooting

If you encounter issues:

  1. Enable debug logging by setting DEBUG=true in your configuration
  2. Check the server logs for error messages
  3. Verify your API key and workspace settings
  4. Ensure your network configuration allows the required connections

For persistent issues, visit the Comet Slack community for support.

Related MCPs

Apple Shortcuts
Developer ToolsJavaScript

Control Apple Shortcuts automations from AI assistants

Clojars Dependency Lookup
Developer ToolsJavaScript

Fetch dependency information from Clojars, the Clojure community's artifact repository

Simple Timeserver
Developer ToolsPython

Provides Claude with current time and timezone information

About Model Context Protocol

Model Context Protocol (MCP) allows AI models to access external tools and services, extending their capabilities beyond their training data.

Generate Cursor Documentation

Save time on coding by generating custom documentation and prompts for Cursor IDE.