An AI assistant for Grafana.
The value is not writing the query. It is going from the query result to the line of code without leaving the session.
Query in natural language
PromQL and LogQL are both easy to read and irritating to compose from memory, especially the rate and histogram_quantile shapes that come up constantly during an incident. Describing what you want and getting the query is a small saving individually and a real one at three in the morning.
The step that matters
Most assistants stop at the query result. This one also has your repository open, so it can find the code that emits a metric or a log line and propose a change. Going from a spike on a dashboard to a diff is the workflow; the query is a step in it rather than the point.
Reads dashboards you already have
It connects to your Grafana rather than replacing it. Existing dashboards, data sources and alert rules remain yours; the agent reads them for context, which means it knows what you already consider worth watching.
Local, because logs contain everything
Log lines routinely include user identifiers, request payloads and occasionally things that should never have been logged. Running against a local model means none of that is transmitted to answer a question about it.
In the Dashboards tab
Questions
What does it need to connect?
A Grafana URL and an API token with read access. Credentials go in the OS keychain rather than a config file.
Can it modify dashboards?
It reads them. Changes to dashboards belong in Grafana, where they are versioned and reviewed like anything else.
Does it work with self-hosted Grafana?
Yes, and that is the common case. Grafana Cloud works too with a token.
Related
- PromQL Query AssistantPromQL is not difficult, it is just impossible to remember the exact shape of a rate over a histogram at the moment you need it.
- Loki Log Analysis with AIFinding the log line is half the job. The other half is working out what emitted it and why.
- AI Design Mode for Web PagesDescribe a page and Design mode returns working HTML you can look at, click through and edit, then implement in the repo you already have open.
- App Wireframe GeneratorLow fidelity on purpose. A wireframe that looks finished gets reviewed as a design instead of as a structure.
Last updated