simplify
Pass
Audited by Gen Agent Trust Hub on Aug 15, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes user-provided code targets, creating a potential surface for indirect prompt injection where malicious instructions embedded in the code could attempt to manipulate the agent's review findings.\n
- Ingestion points: The skill ingests a 'target supplied by the user' or code from an 'upstream merge-base diff' (SKILL.md).\n
- Boundary markers: The instructions provide clear functional boundaries ('read-only maintainability review') but do not implement technical delimiters or specific instructions to ignore embedded prompts within the code targets.\n
- Capability inventory: The skill is strictly limited to read-only analysis and is explicitly prohibited from changing code, tests, documentation, or state (SKILL.md).\n
- Sanitization: There is no evidence of specific sanitization or filtering logic applied to the external code targets before they are processed by the agent.
Audit Metadata