tailwind-design-system
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes local project files and source material, creating an attack surface where instructions embedded in those files could influence agent behavior. 1. Ingestion points: Inspection of available files, runtime, versions, and inputs in
SKILL.md. 2. Boundary markers: There are no explicit instructions for the agent to use delimiters or ignore embedded instructions within the source material. 3. Capability inventory: The skill is authorized to implement artifacts and call thesandbase_runandsandbase_discovertools. 4. Sanitization: No sanitization or validation logic is defined for data extracted from project files.
Audit Metadata