ui-design-system
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a local Python script
scripts/design_token_generator.pyto perform design calculations. This script is bundled with the skill and executed by the agent to generate JSON, CSS, and SCSS outputs. This activity is restricted to the local environment and the skill's specific purpose. - [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing user-provided inputs like hex color codes and style parameters. 1. Ingestion points: Inputs are accepted in the brand color and style preference parameters in SKILL.md. 2. Boundary markers: None present. 3. Capability inventory: The skill instructions involve executing a Python script via shell and redirecting output to files. 4. Sanitization: The Python script uses the
argparselibrary for argument parsing and validation, andcolorsysfor mathematical color space conversion, which effectively constrains user input to expected formats. - [SAFE]: No malicious obfuscation, persistence, data exfiltration, or dangerous command execution patterns were detected. The skill scripts and documentation are consistent with the primary purpose of UI design system management.
Audit Metadata