ui-ux-pro-max
Pass
Audited by Gen Agent Trust Hub on Jun 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates primarily by performing keyword-based searches against bundled CSV data files using a custom implementation of the BM25 algorithm.
- [COMMAND_EXECUTION]: The skill includes a helper script (
design_system.py) that the agent is encouraged to run. This script is well-structured, uses standard Python libraries, and its execution is scoped to the skill's data processing tasks. - [DATA_EXPOSURE]: No access to sensitive system paths or credentials was identified. The file persistence functionality in
design_system.pyis used to save generated design specifications into a localdesign-system/directory within the user's project, which is consistent with its stated purpose. - [REMOTE_CODE_EXECUTION]: There are no external downloads, piped shell commands, or attempts to install unverified dependencies.
- [INDIRECT_PROMPT_INJECTION]: While the skill ingests user queries to search its database, it treats the results as informational data for the agent to reference rather than instructions to execute. The risk is minimized by the use of static, bundled asset files.
Audit Metadata