home / mcp / agentops mcp server

AgentOps MCP Server

agentops mcp

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "agentops-ai-agentops-mcp": {
      "command": "npx",
      "args": [
        "agentops-mcp"
      ]
    }
  }
}

The AgentOps MCP server exposes observability and tracing data to help you debug complex AI agent runs. It provides context about where an agent succeeds or fails, enabling faster troubleshooting and performance improvements.

How to use

Configure your MCP client to connect to the AgentOps MCP server by using the stdio configuration. This server runs locally as a process you start from your development environment, allowing you to request traces and spans during agent runs.

How to install

Prerequisites you need before installing and running the MCP server:

- Node.js version 18.0.0 or higher

Install and run the MCP server locally using the following steps.

# Clone the MCP project repository
git clone https://github.com/AgentOps-AI/agentops-mcp.git
cd agentops-mcp

# Install dependencies
npm install

# Build the project
npm run build

# Pack the server (prepare for local execution in your environment)
npm pack

Configuration and usage notes

Use the following MCP client configuration to run the AgentOps MCP server as a stdio process. The server is started via a simple command that launches the MCP package and exposes the required tools.

{
  "mcpServers": {
    "agentops_mcp": {
      "command": "npx",
      "args": ["agentops-mcp"],
      "env": {
        "AGENTOPS_API_KEY": ""
      }
    }
  }
}

Available tools

auth

Authorize using an AgentOps project API key and return a JWT token.

get_trace

Retrieve trace information by ID.

get_span

Get span information by ID.

get_complete_trace

Get comprehensive trace information including all spans and their metrics.