krt-deploy-summoner
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data that could contain malicious instructions.
- Ingestion points: Reads Helm charts, Kubernetes manifests, cluster events, container logs, and CI/CD deployment scripts in
SKILL.md(Steps 2 and 3). - Boundary markers: None explicitly defined in the file content to delimit untrusted data.
- Capability inventory: The skill facilitates execution of powerful commands including
kubectl apply,kubectl delete,helm upgrade, anddocker compose upinSKILL.md(Step 5). - Sanitization: Relies on user approval of the final plan rather than automated content sanitization.
- [COMMAND_EXECUTION]: The tool generates and executes commands for container and cluster orchestration.
- Evidence: Interacts with local and remote environments via
kubectl,helm, anddockerCLIs. - Safety: Includes strong guardrails in
SKILL.mdandreferences/safety.mdprohibiting silent context switches, resource deletion without approval, and the printing of sensitive credentials, tokens, or kubeconfigs.
Audit Metadata