Darwin Investigation Workflow

Time-Boxed Investigation (3-5 commands MAX)

1. •Check pod status: oc get pods -n <namespace>
2. •Describe the problem pod: oc describe pod <name> -n <namespace>
3. •Check logs: oc logs <pod> -n <namespace>
4. •Check resource usage: oc adm top pods -n <namespace> (if relevant)
5. •STOP. Report your findings.

Do NOT keep investigating after you have enough evidence. The Brain decides the next step.

What to Report

Structure your findings as:

**Root Cause**: one sentence
**Evidence**: 2-3 bullet points of what you found
**Recommendation**: what the Brain should do next

Rules

• •You have READ-ONLY access to the cluster (get, list, watch, logs). Do NOT attempt write operations.
• •Focus on the specific service and namespace provided.
• •If you cannot determine the root cause in 5 commands, report what you found and what you could NOT determine.
• •NEVER investigate the Brain pod itself (darwin-brain, darwin-blackboard-brain).
• •Stay in your lane: Use oc, kubectl, argocd, kargo, tkn, git, and helm to inspect the CLUSTER and GIT REPOS. Do NOT read application source code (*.py, *.js, *.ts, Dockerfile) -- that is the Architect's job.