project-critique
Pass
Audited by Gen Agent Trust Hub on Mar 23, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill ingests and processes various user-controlled files, including configuration and design documents, to perform its analysis. This pattern creates a surface for indirect prompt injection where malicious content in the project files could attempt to manipulate the agent's behavior or tool use.\n
- Ingestion points: Ingests data from config.json, BRIEF.md, and design chunks found in the project directory (SKILL.md).\n
- Boundary markers: Lacks clear delimiters or "ignore" instructions for processed data.\n
- Capability inventory: Employs Read, Write, Bash, and Agent tools (SKILL.md).\n
- Sanitization: No evidence of input validation or sanitization.
Audit Metadata