azsdk-common-apiview-feedback-resolution

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill is designed for resolving API review feedback on Azure SDK PRs using official tools from the azure-sdk-mcp server. The tools and behaviors described align with the stated purpose and the author's identity as Azure.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it processes external reviewer comments to drive code changes. 1. Ingestion point: Reviewer comments are retrieved from APIView URLs using the azsdk_apiview_get_comments tool as described in SKILL.md. 2. Boundary markers: There are no explicit instructions or delimiters defined to isolate or sanitize instructions embedded within the reviewer comments. 3. Capability inventory: The skill has the capability to modify the local filesystem via azsdk_customized_code_update, regenerate code via azsdk_package_generate_code, and create GitHub issues via azsdk_typespec_delegate_apiview_feedback (defined in SKILL.md). 4. Sanitization: No sanitization or content validation for the fetched comments is documented in the skill logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 08:08 AM
Security Audit — agent-trust-hub — azsdk-common-apiview-feedback-resolution