Coval MCP Server

The official Model Context Protocol server for Coval - the AI evaluation platform.

This MCP server allows AI assistants like Claude Desktop and Cursor to interact with Coval's evaluation APIs, enabling you to:

• Launch and monitor evaluation runs
• Manage AI agents and test sets
• Retrieve evaluation metrics and results

Quick Start

Claude Desktop Setup

1. Get your Coval API key from dashboard.coval.dev

2. Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "coval": {
      "command": "npx",
      "args": ["-y", "@coval/mcp-server"],
      "env": {
        "COVAL_API_KEY": "your_api_key_here"
      }
    }
  }
}

3. Restart Claude Desktop

Remote Connection (Recommended for Production)

{
  "mcpServers": {
    "coval": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://mcp.coval.dev/mcp",
        "--header",
        "X-API-KEY: ${COVAL_API_KEY}"
      ],
      "env": {
        "COVAL_API_KEY": "your_api_key_here"
      }
    }
  }
}

Available Tools

Runs

• list_runs - List evaluation runs with filtering
• get_run - Get details of a specific run (includes metrics for completed runs)
• create_run - Launch a new evaluation run
• delete_run - Cancel or delete a run

Agents

• list_agents - List configured agents
• get_agent - Get agent configuration
• create_agent - Create a new agent
• update_agent - Update agent configuration

Test Sets

• list_test_sets - List available test sets
• get_test_set - Get test set details
• create_test_set - Create a new test set

Metrics

• list_metrics - List available metric definitions
• get_metric - Get metric details

Personas

• list_personas - List simulated personas
• get_persona - Get persona details

Example Usage

Once connected, you can ask Claude things like:

"Show me my recent evaluation runs"

"List all my agents"

"Run an evaluation of my customer-support-agent against the billing-inquiries test set"

"What are the metrics for run abc123?"

Development

# Install dependencies
npm install

# Build
npm run build

# Test locally with MCP Inspector
npm run inspector

# Run tests
npm test

Environment Variables

| Variable | Required | Default | Description | |----------|----------|---------|-------------| | COVAL_API_KEY | Yes | - | Your Coval API key | | COVAL_API_BASE_URL | No | https://api.coval.dev/v1 | API base URL | | LOG_LEVEL | No | info | Logging level |

Documentation

• Coval Documentation
• MCP Protocol
• API Reference

License

MIT

Support

• GitHub Issues
• Coval Support