Skip to content

mcp-integration.md

Latest commit

 

History

History
97 lines (67 loc) · 3.07 KB

mcp-integration.md

File metadata and controls

97 lines (67 loc) · 3.07 KB
Raw

AI Tool Integration (MCP)

The grafana/otel-lgtm image integrates with Model Context Protocol (MCP) so AI coding tools (Claude, Cursor, etc.) can query your telemetry data directly.

Tempo exposes an HTTP MCP endpoint from inside the container. Grafana data is accessed via a client-side uvx mcp-grafana process that runs on your machine and connects to the Grafana instance in the container.

What you get

  • Dashboards: list, read, and search dashboards via client-side uvx mcp-grafana
  • Logs: query via LogQL via client-side uvx mcp-grafana
  • Metrics: query via PromQL via client-side uvx mcp-grafana
  • Traces: query via TraceQL via client-side uvx mcp-grafana or through Tempo's built-in HTTP MCP endpoint (in-container)

Setup

  1. Enable the Tempo MCP server:

    echo 'TEMPO_EXTRA_ARGS=--query-frontend.mcp-server.enabled=true' >> .env

  2. Start the container:

    ./run-lgtm.sh

  3. Get the MCP configuration:

    docker exec lgtm cat /etc/lgtm/mcp.json   # or: podman exec ...

  4. Paste the JSON into your AI tool's MCP configuration.

    For Claude Code, you can add the servers individually:

    # Get the service account token
TOKEN=$(docker exec lgtm cat /tmp/grafana-sa-token)   # or: podman exec ...

# Add the Grafana MCP server (requires uvx)
claude mcp add grafana \
  -e GRAFANA_URL=http://localhost:3000 \
  -e GRAFANA_SERVICE_ACCOUNT_TOKEN="$TOKEN" \
  -- uvx mcp-grafana

    The Grafana MCP server proxies the Tempo MCP server, but you can add it separately if you wish:

    # Add the Tempo MCP server
claude mcp add --transport http tempo http://localhost:3200/api/mcp

Backend mapping

Component MCP Server Transport What you can query
Grafana grafana stdio Dashboards, PromQL, LogQL
Tempo tempo HTTP Traces via TraceQL

Collector debug exporter

The OpenTelemetry Collector includes a debug exporter that logs all received telemetry to stdout. This is useful for verifying that data is flowing correctly.

Enable it by setting the environment variable before starting the container:

OTEL_COLLECTOR_DEBUG_EXPORTER=true ./run-lgtm.sh

This adds the debug exporter to the logs, metrics, and traces pipelines. The output appears in the collector's logs (enable with ENABLE_LOGS_OTELCOL=true or ENABLE_LOGS_ALL=true).

OBI (eBPF auto-instrumentation)

When OBI is enabled, it generates traces and RED metrics automatically. These are queryable via PromQL through the Grafana MCP server:

# Number of instrumented processes
obi_instrumented_processes

# HTTP request duration (RED metrics)
http_server_request_duration_seconds_count{http_route="/rolldice"}

See the OBI section in the README for setup instructions.