ui-design-system
Pass
Audited by Gen Agent Trust Hub on Jul 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a local Python script (
scripts/design_token_generator.py) to automate the generation of design tokens. This involves the AI agent executing shell commands with arguments derived from user input such as hex color codes and style preferences. This is standard functionality for the skill's purpose.\n- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by design, as it interpolates user-provided values directly into shell command templates for token generation. While no malicious instructions were found in the static content, the ingestion of untrusted data into command arguments represents a potential attack surface.\n - Ingestion points: User-provided strings for brand primary colors and style preferences in
SKILL.mdworkflows.\n - Boundary markers: Not present; the workflows use direct interpolation without delimiters or instructions to ignore embedded commands.\n
- Capability inventory: Execution of a local Python script via the shell, with the ability to write output to the local file system (e.g.,
.css,.jsonfiles).\n - Sanitization: None; the skill does not explicitly instruct the agent to validate or sanitize the input strings before they are used as command-line arguments.
Audit Metadata