k8s-incident
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes potentially attacker-controlled data from Kubernetes clusters which could lead to unauthorized actions if malicious instructions are embedded in that data.
- Ingestion points: The skill reads pod logs (
kubectl_logs), cluster events (kubectl_get), and resource descriptions (kubectl_describe) inSKILL.mdandscripts/collect-diagnostics.py. - Boundary markers: No specific delimiters or instructions to ignore embedded commands are used when processing the collected data.
- Capability inventory: The agent has access to management tools such as
delete_resource,kubectl_rollout, andhelm_rollback. - Sanitization: The
scripts/collect-diagnostics.pyscript aggregates data without sanitizing or escaping content before it is returned to the agent context.
Audit Metadata