first-principles-review
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external data, including project files, logs, and third-party requirements, which creates a potential surface for indirect prompt injection. Maliciously crafted content in these data sources could attempt to influence the agent's architectural judgment. \n
- Ingestion points: SKILL.md (specifies reading project files, tests, logs, and requirements). \n
- Boundary markers: Absent. \n
- Capability inventory: No script execution or network tools defined within this skill. \n
- Sanitization: Absent.
Audit Metadata