write-a-skill
Pass
Audited by Gen Agent Trust Hub on Aug 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill transforms user-supplied requirements into agent skills, creating a vector for indirect prompt injection.
- Ingestion points: External requirements are ingested from conversation context or handoff documents (SKILL.md).
- Boundary markers: The process lacks clear delimiters or instructions to ignore malicious content within the ingested requirements.
- Capability inventory: The skill facilitates file creation and modification (SKILL.md, REFERENCE.md, and utility scripts).
- Sanitization: No explicit sanitization or validation of the input requirements is performed before generating skill content.
Audit Metadata