What it does
You describe what is wrong and it investigates across all three signal sources at once: reading cluster state through read-onlykubectl, pulling the metrics, logs and traces around the incident
from Grafana, and checking GPU health where the workload depends on it. It reconstructs the change
timeline and returns a structured root-cause analysis.
The point of combining them is correlation that neither half can do alone — a pod restarting because
of a GPU fault looks like an ordinary crash loop until something reads both.
What it reads
All of it read-only, apart from the in-pod diagnostics noted below.
It carries the same domain knowledge as Kubernetes
RCA — twenty-three workload-specific methods —
plus five GPU-specific ones.
Before you deploy
The Grafana connection is optional because the agent can instead reach your own in-cluster MCP
server, or run its own bundled Grafana tooling. Add a connection only to use the built-in path.
Without any of the three, it still investigates the cluster — it simply has no telemetry to
correlate against.
view role, a
supplemental role for the cluster-scoped reads it omits, and pods/exec for the workload methods
that diagnose from inside the pod. Exec is a separate value, so you can narrow it to named
namespaces or remove it and let those methods fall back to resource state.
Ask it for
In chat, or as a run’s prompt:Defaults and limits
The time budget degrades rather than fails. At 70% the agent is nudged to converge; at 100% further
tool calls are refused and it reports what it has.
In a workflow
It is a specialist bound to one cluster, so a step targeting it has already established which cluster the problem is in. Because it covers three signal sources, it often replaces two steps that would otherwise run in parallel and need synthesizing afterwards. See Orchestration for how a step delivers work to it.What it will not do
It changes nothing in the cluster or in Grafana. Its non-read grant is confined to a fixed list of read-only in-pod diagnostics, and everything else is refused. It sees exactly one cluster. Deploying one instance and expecting it to answer for a fleet is the mistake worth avoiding — run one per cluster, and name each after the cluster it watches.Next steps
Kubernetes RCA
The cluster half alone, with no external data sources.
Grafana Investigator
The telemetry half alone, across clusters.
Deploy an agent
Install a self-hosted worker into your cluster.
Agent catalog
Every catalog agent, side by side.