agency-inclusive-visuals-specialist
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied creative briefs to generate output, which presents a surface for indirect instructions to influence agent behavior.
- Ingestion points: In SKILL.md, 'Phase 1: The Brief Intake' indicates the agent analyzes requested creative briefs.
- Boundary markers: The skill lacks explicit technical boundary markers or delimiters for untrusted input in SKILL.md, though it uses a structured 'Annotation Framework' to organize output.
- Capability inventory: In SKILL.md, the skill is limited to text prompt generation and lacks high-risk capabilities such as file system writes, network requests, or shell execution.
- Sanitization: The 'Phase 4: The Review Gate' in SKILL.md provides a 7-point QA checklist for human verification of assets, which serves as a manual sanitization step.
Audit Metadata