judge
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes 'Completed work output (files, changes, artifacts)' as primary input to perform evaluations.
- Ingestion points:
SKILL.mddefines work output, files, and artifacts as primary inputs for the review process. - Boundary markers: The instructions lack explicit delimiters or specific warnings to ignore instructions found within the reviewed artifacts, which could lead the agent to follow embedded malicious commands.
- Capability inventory: The skill is intended to review code changes, document generation, and infrastructure updates, requiring deep inspection of file contents.
- Sanitization: No sanitization or filtering of input content is specified before the evaluation process.
Audit Metadata