ui-ux-pro-max
Fail
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: HIGHPRIVILEGE_ESCALATIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PRIVILEGE_ESCALATION]: The skill instructs the agent to execute
sudo apt update && sudo apt install python3on Linux systems. Requesting administrative privileges viasudois a high-risk pattern that allows for system-wide modifications. - [COMMAND_EXECUTION]: The workflow relies on executing a local Python script
skills/ui-ux-pro-max/scripts/search.py. The actual content of this script is referenced through an external relative path (../../../src/ui-ux-pro-max/scripts), meaning the code is not contained within the skill package itself and cannot be audited locally. - [INDIRECT_PROMPT_INJECTION]: The skill implements a 'Master + Overrides' pattern where it reads from external Markdown files to generate design recommendations.
- Ingestion points:
design-system/MASTER.md,design-system/pages/*.md, andui-reasoning.csv. - Boundary markers: Absent. The instructions explicitly tell the agent to 'prioritize' and 'exclusively' follow rules found in these files without safety delimiters or instructions to ignore embedded commands.
- Capability inventory: The skill uses
python3to execute scripts, creates directories, and persists design system files to the local disk. - Sanitization: No sanitization or validation of the ingested file content is performed before processing.
Recommendations
- AI detected serious security threats
Audit Metadata