Agent Evals by Galileo

official

Bring agent evaluations, observability, and synthetic test set generation directly into your IDE for free with Galileo's new MCP server

What can you do with Agent Evals By Galileo MCP?

  • Create synthetic datasets — Ask your assistant to generate datasets of queries, including prompt injection, off-topic, or toxic content scenarios.
  • Check dataset status — Track dataset generation progress and preview the first 10 rows of generated data.
  • Build prompt templates — Create reusable prompt templates with model configurations and temperature settings for consistent behavior.
  • Get experiment guidance — Ask how to set up and run Galileo experiments, including dataset prep and metrics configuration.
  • Analyze log stream signals — Fetch recent signals from your Log streams and get suggested fixes for issues found.
  • Get integration help — Request step-by-step guides for adding Galileo observability to OpenAI or LangChain applications.

Documentation

Galileo MCP Server

Learn how to integrate Galileo's Model Context Protocol (MCP) server with AI-enabled IDEs like Cursor and VS Code

The Galileo Model Context Protocol (MCP) server enables seamless integration between AI-powered IDEs, such as Cursor, or VS Code with GitHub Copilot, and Galileo's evaluation and observability platform.

With MCP, you can access Galileo's capabilities directly from your development environment, including:

  • Creating and managing datasets
  • Running experiments
  • Setting up prompt templates
  • Getting signals on Log streams
  • Integrating Galileo with your code

Prerequisites

Before you begin, ensure you have the following:

1. AI-enabled IDE

Install an AI-enabled IDE such as Cursor or VS Code with AI capabilities

2. API key

Generate your Galileo API key from the API keys page

Configure your IDE

The Galileo MCP server works with both Cursor and VS Code. Follow the steps below for your IDE:

[tab] VS Code

1. Install GitHub Copilot

Install the GitHub Copilot extension if you haven't already

2. Open MCP settings

Open the Command Palette (Ctrl + Shift + P on Windows/Linux, or Cmd + Shift + P on Mac) and search for "MCP: Open User Configuration"

3. Add the Galileo MCP server configuration

Copy and paste the configuration below. Replace YOUR-API-KEY with your actual Galileo API key.

{
  "servers": {
    "galileo_mcp_server": {
      "url": "https://api.galileo.ai/mcp/http/mcp",
      "headers": {
        "Galileo-API-Key": "YOUR-API-KEY",
        "Accept": "text/event-stream"
      }
    }
  },
  "inputs": []
}

This assumes you are using app.galileo.ai.

If you're using a self-hosted Galileo deployment, replace the https://api.galileo.ai/mcp/http/mcp url with your deployment URL. The format of this URL is based on your console URL, replacing console with api and appending /mcp/http/mcp.

For example:

  • if your console URL is https://console.galileo.example.com, the MCP url would be https://api.galileo.example.com/mcp/http/mcp
  • if your console URL is https://console-galileo.apps.mycompany.com, the MCP url would be https://api-galileo.apps.mycompany.com/mcp/http/mcp

4. Reload VS Code

Reload VS Code by opening the Command Palette and running "Developer: Reload Window" for the changes to take effect

[tab] Cursor

1. Open the Tools & MCP settings

Open the Cursor command palette (Ctrl + Shift + P on Windows/Linux, or Cmd + Shift + P on Mac), then choose "Open MCP Settings".

2. Open the MCP server configuration file

Click on Add Custom MCP to open the mcp.json MCP configuration file, or New MCP Server if you already have other MCP servers configured.

3. Add the Galileo MCP server configuration

Copy and paste the configuration below. Replace YOUR-API-KEY with your actual Galileo API key.

{
  "mcpServers": {
    "galileo_mcp_server": {
      "url": "https://api.galileo.ai/mcp/http/mcp",
      "headers": {
        "Galileo-API-Key": "YOUR-API-KEY",
        "Accept": "text/event-stream"
      }
    }
  }
}

This assumes you are using app.galileo.ai.

If you're using a self-hosted Galileo deployment, replace the https://api.galileo.ai/mcp/http/mcp url with your deployment URL. The format of this URL is based on your console URL, replacing console with api and appending /mcp/http/mcp.

For example:

  • if your console URL is https://console.galileo.example.com, the MCP url would be https://api.galileo.example.com/mcp/http/mcp
  • if your console URL is https://console-galileo.apps.mycompany.com, the MCP url would be https://api-galileo.apps.mycompany.com/mcp/http/mcp

4. Save and restart

Save the configuration and restart Cursor for the changes to take effect

[!TIP]

The configuration is the same for both Cursor and VS Code. Make sure to replace YOUR-API-KEY with your actual Galileo API key from the API keys page.

Verify your setup

Once configured, you can verify your MCP setup by asking your AI assistant in your IDE:

Can you show me how to add Galileo logging to my agent bot?
Help me create a synthetic dataset for customer support queries
How do I integrate Galileo with LangChain?

Your AI assistant should now be able to access Galileo's capabilities and respond with information from your Galileo account.

Tools

The Galileo MCP server provides powerful tools that you can access through natural conversation with your AI assistant. Simply ask questions or make requests, and the AI will use these tools to help you.

[accordion] Create Datasets

Generate synthetic datasets or upload your own data to test and evaluate your AI applications. The tool supports creating datasets with various types of queries including general queries, prompt injections, off-topic content, and toxic content scenarios.

What you can ask:

Create a dataset with 50 customer service queries about billing issues
Generate a dataset of 30 chatbot queries, including some prompt injection attempts
Make a dataset with product recommendation queries and include off-topic questions

[accordion] Check Dataset Status

Track the progress of your dataset generation and preview the generated content. You'll see the first 10 rows of data along with generation status and progress updates.

What you can ask:

Check the status of my dataset that's currently generating
Show me the preview of dataset abc-123
Is my customer service dataset ready yet?

[accordion] Create Prompt Templates

Build reusable prompt templates that you can use across all your projects. Set up model configurations, temperature settings, and other parameters for consistent prompt behavior.

What you can ask:

Create a prompt template called "Friendly Assistant" for customer support responses
Make a prompt template for summarizing technical documentation with lower temperature
Set up a chat template for a code review assistant

[accordion] Set Up Experiments

Get complete guidance on setting up and running Galileo experiments, including dataset preparation, metrics configuration, and integration with your existing code. Available for Python and other supported languages.

What you can ask:

How do I set up a Galileo experiment in Python?
Show me how to run an experiment with my RAG application
Guide me through creating an experiment for my agentic workflow

[accordion] Get Signals

Analyze your application's Log streams to identify issues, patterns, and opportunities for improvement. Get specific recommendations based on your logged data.

What you can ask:

What issues do you see in my production log stream?
Fetch the most recent signals from Galileo and propose fixes for them in my code
Get insights about my chatbot log stream and suggest improvements

[accordion] Integrate with OpenAI

Get step-by-step integration guides for adding Galileo observability to your OpenAI applications. Automatically log prompts, responses, model parameters, and token usage with minimal code changes.

What you can ask:

How do I add Galileo logging to my OpenAI application?
Show me how to integrate Galileo with OpenAI in TypeScript
Help me set up Galileo observability for my GPT-4 chatbot

[accordion] Integrate with LangChain

Get complete integration instructions for adding Galileo to your LangChain applications. Capture full traces of your chains, agents, and tools with automatic logging.

What you can ask:

How do I integrate Galileo with my LangChain application?
Show me how to add Galileo tracing to my LangChain agent
Help me log my LangChain RAG pipeline with Galileo

[accordion] Search Documentation

Find relevant information, code examples, API references, and implementation guides across all Galileo documentation. Get direct links to the pages you need.

What you can ask:

How do I set up data logging in Galileo?
Find documentation about custom metrics
Search for examples of agentic AI evaluation

Example use cases

Create a synthetic dataset

Ask your AI assistant:

Create a synthetic dataset with 20 customer service queries about product
returns. Include both general queries and off-topic queries.

The MCP server will guide you through the dataset creation process and provide a dataset ID to track progress.

Get integration help

Ask your AI assistant:

How do I integrate Galileo with my LangChain application in Python?

The MCP server will provide complete integration code examples and setup instructions.

Get Signals

Ask your AI assistant:

Fetch the most recent signals from Galileo and propose fixes for them in my code

The MCP server will analyze your Log stream and suggest improvements.

Troubleshooting

[accordion] Error connecting to Galileo MCP Server

  • Check that the MCP server URL is set to https://api.galileo.ai/mcp/http/mcp
  • Ensure the Accept header is set to text/event-stream
  • Restart your IDE MCP connection after making configuration changes

[accordion] API key errors

  • Confirm your API key is properly set in the configuration
  • Check that your API key has not expired
  • Generate a new API key from the API keys page

Next steps

[card] Log your first trace

Learn how to log your first trace with Galileo

[card] Run your first experiment

Set up and run experiments to evaluate your AI applications

[card] Explore integrations

Learn about Galileo integrations with third-party frameworks