09-design-systems
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill is strictly instructional and does not package any executable code or scripts.\n- [SAFE]: No evidence of prompt injection, data exfiltration, or persistence mechanisms was found. The instructions align with professional engineering standards.\n- [INDIRECT_PROMPT_INJECTION]: The skill involves processing data from the local repository to understand existing frameworks and conventions (SKILL.md).\n
- Ingestion points: Local repository content (files, dependencies, framework configuration).\n
- Boundary markers: None provided in the instructions.\n
- Capability inventory: File system read access and code generation capabilities.\n
- Sanitization: None specified.\n
- Assessment: The requirement to inspect the repository is a standard function for this agent use case and does not introduce an elevated risk level.
Audit Metadata