cloud-architect
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the repository (such as configuration files, manifests, and dependency lists) to generate its deployment recommendations.
- Ingestion points: The agent reads "dependency manifests, entry points, and configuration" along with "Dockerfiles, Compose files, Kubernetes manifests, IaC, [and] CI deploy jobs" in Step 1.
- Boundary markers: The instructions do not specify any boundary markers or instructions to treat data from the repository as untrusted or to ignore embedded natural language instructions within those files.
- Capability inventory: The skill allows the agent to read arbitrary files within the repository and write a new file (
deploy-plan.md). - Sanitization: There is no evidence of sanitization, validation, or escaping of the repository content before it is interpolated into the agent's context or used to generate the final plan.
Audit Metadata