MCP server · Analytics

Spoor MCP server

Spoor tells your AI what changed in your services and where a failure started, using metrics, logs and traces.

Analytics For it and development For operations Monitoring

Spoor is a small helper program that connects your AI to the VictoriaMetrics monitoring stack (VictoriaMetrics, VictoriaLogs and VictoriaTraces). When something breaks, you ask your AI what changed, and Spoor gives back a short, ranked answer that points to the service where the problem actually started. It is handy for people who look after online services and use an AI assistant to help figure out why something stopped working.

What is an MCP server? The 30-second version

On its own, your AI can only chat. It cannot see your monitoring data. An MCP server is a small helper program that gives your AI a new skill or a connection to another tool. Spoor is that helper here. It connects your AI to your VictoriaMetrics, VictoriaLogs and VictoriaTraces data, so the AI can look things up there and answer your questions about what is going wrong.

What this MCP server does

You ask your AI something like what changed in the last 30 minutes. The AI passes that question to Spoor. Spoor then queries your VictoriaMetrics, VictoriaLogs and VictoriaTraces backends, compares the recent window against a baseline, and looks at trace trees to see where failures started. It ranks the changes, separates the real cause from services that are just suffering as a result, and gives back a short digest. Your AI then explains that digest to you in plain language.

What you can do with it

Ask what changed in the last 30 minutes across all your services
Find out which service a failure started in, not just which one is loudest
See new or spiking error log patterns with counts and an example
Spot spans that got slower or started failing
List the services in a time window and which signals have data
Check errors that were already there so you do not chase them
Keep the answer short by setting a limit on how much text comes back

What it gives back to you

You get a short summary sentence and a ranked list of findings. Each finding names a service, says which signal it came from (metrics, logs or traces), gives a score, and describes what changed in plain words. Some findings include a link back to the query behind them so you can check the work. There is also a coverage line that honestly says when a backend had no data or failed to answer.

Before you start

What you need

Go 1.25 or newer to build it, or a prebuilt version from the project
Access to a VictoriaMetrics, VictoriaLogs and VictoriaTraces setup
The addresses of those three services (they look like web links)
Docker if you want to try the built-in demo data

Good to know

Spoor is read-only and never changes your data, but it does read your logs and traces, so make sure you are allowed to send that data to your AI assistant.

Install it with your AI

Add Spoor MCP server to your AI, no technical skills needed

You don't install anything by hand. You copy one prompt, paste it into an AI that can work on your computer, and it checks, installs and connects the server for you, asking you when it needs something.

Sign in to get the install prompt

Members get a ready-made prompt that lets the Claude desktop app check Spoor MCP server, install it and connect it for them, step by step. You don't need any technical skills: you copy, paste and answer a few questions. Your connected AI can also find and install any of the 4,361 MCP servers here for you.