sip
Pass
Audited by Gen Agent Trust Hub on Jul 15, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill instructs the agent to analyze artifacts or other skills created during the session, which serves as an ingestion point for untrusted data.
- Ingestion points: Artifacts and skill files identified in the verification workflow (SKILL.md).
- Boundary markers: None specified to differentiate data content from instructions within the processed artifacts.
- Capability inventory: The skill orchestrates multiple tools (e.g., shower, factchk, mandela) to analyze the input.
- Sanitization: No explicit sanitization or filtering of the input content is described.
- [NO_CODE]: The skill consists entirely of instructional markdown and does not include any scripts, binaries, or executable components.
Audit Metadata