home / mcp / pydantic logfire mcp server
Provides a hosted MCP server to access OpenTelemetry traces and metrics from Pydantic Logfire.
Configuration
View docs{
"mcpServers": {
"pydantic-logfire-mcp": {
"url": "https://logfire-us.pydantic.dev/mcp"
}
}
}You can access OpenTelemetry traces and metrics from Pydantic Logfire through a hosted MCP server or run a local MCP server. This MCP server enables you to connect your tracing and metrics data to your tools and clients, simplifying integration and observability workflows.
You connect your MCP clients to the remote Pydantic Logfire MCP server to start sending and querying traces and metrics. Choose the region that matches your data, authenticate when prompted, and configure your client with the provided MCP endpoint. This approach requires no local server setup.
There is a hosted MCP server you can use directly by configuring your MCP client. No local server installation is required when you use the remote endpoint. Follow these steps to configure your client for the US region and EU region.
{
"mcpServers": {
"logfire": {
"type": "http",
"url": "https://logfire-us.pydantic.dev/mcp"
}
}
}{
"mcpServers": {
"logfire": {
"type": "http",
"url": "https://logfire-eu.pydantic.dev/mcp"
}
}
}Notes and practical guidance for using the remote MCP endpoints are provided below. You will typically create a configuration in your MCP client that points to the remote URL, then initiate the connection. When you connect for the first time, you will be prompted to authenticate with your Pydantic Logfire account through your browser.
Choose the region that aligns with where your data resides to minimize latency. Authentication is handled through your browser for the remote server, so ensure you are signed into the correct account before connecting. If you need to run a local MCP server for development or testing, a separate local setup guide exists outside this immediate workflow.