ui-design-system
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill workflow involves passing user-supplied inputs, such as color hex codes and design preferences, into shell commands to execute a local Python script. This interpolation creates a surface for command injection if the agent's execution environment does not adequately sanitize user input.
- Ingestion points: User-provided color values and style preferences identified in the "Generate Design Tokens" workflow of SKILL.md.
- Boundary markers: Shell commands are wrapped in markdown code blocks within SKILL.md and reference files.
- Capability inventory: The skill utilizes the agent's ability to execute local scripts (scripts/design_token_generator.py) and write output to the filesystem via shell redirection (documented in SKILL.md).
- Sanitization: The provided Python script performs minimal input processing (e.g., stripping the '#' from hex codes) but lacks robust validation to prevent shell injection via other arguments.
Audit Metadata