azure-review

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill's Output Format explicitly asks for "Evidence: " and its process/prerequisites require scanning IaC and Key Vault-related files (which may contain secrets), but it gives no guidance to redact or avoid verbatim reproduction, creating a likely need for the LLM to include secret values in its output.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.70). SKILL.md instructs the runtime to discover and read Azure IaC/config files from the repository (e.g., Terraform/Bicep/ARM/policy JSON), and those files can contain outsider-authored free text (comments/descriptions/string values) that the agent may ingest into the LLM context for evaluation.