critique
Pass
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill identifies a surface for indirect prompt injection by ingesting untrusted data for review.\n
- Ingestion points: SKILL.md Phase 1 identifies scope from user arguments, file changes, and conversation history.\n
- Boundary markers: None detected; the prompts do not explicitly instruct the sub-agents to ignore instructions contained within the reviewed content.\n
- Capability inventory: The skill uses the
Tasktool to orchestrate parallel reviews across three specialized agents.\n - Sanitization: None detected; the content is passed directly to the review agents.
Audit Metadata