az

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill includes commands that embed secrets directly (e.g., --value "SuperSecret123!" and --admin-password placeholders) and provides a literal example secret, which requires or encourages the LLM to handle/output secret values verbatim.

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

• Third-party content exposure detected (high risk: 0.70). The skill contains commands that fetch and deploy code/artifacts from arbitrary public URLs and registries (e.g., az repos import / az webapp deployment source config --repo-url, az acr import --source docker.io/..., az storage blob copy --source-uri), which are untrusted/user-provided sources that would be fetched and interpreted as part of workflows and could enable indirect prompt injection.