pack-prerequisite-checker
Pass
Audited by Gen Agent Trust Hub on Nov 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill acts as a logic gate and routing mechanism with no power to execute code or access sensitive data. It explicitly prohibits arbitrary interfaces and enforces official runtimes.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user data (briefs) which could contain malicious instructions. However, the risk is negligible as the skill has no tools or actions to exploit. Evidence chain: 1. Ingestion points: Current brief summary in SKILL.md and current_brief_status in contract.json. 2. Boundary markers: None identified in references or instructions. 3. Capability inventory: No tool usage, subprocess calls, or network operations defined in any file. 4. Sanitization: None identified in instructions.
Audit Metadata