ui-ux-pro-max-v3
Pass
Audited by Gen Agent Trust Hub on Jul 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for the user to execute shell commands to perform design searches and generate reports. It also includes instructions for system package management that require elevated privileges.
- Evidence in
SKILL.md:python3 .claude/skills/ui-ux-pro-max/scripts/search.py ...,sudo apt update && sudo apt install python3. - [EXTERNAL_DOWNLOADS]: The skill directs the user to download and install system dependencies (Python 3) using official OS package managers. This is documented for environment setup.
- Evidence in
SKILL.md:brew install python3,winget install Python.Python.3.12. - [PROMPT_INJECTION]: The search functionality processes user-provided strings as input for a BM25 ranking engine. This represents a potential surface for indirect prompt injection where malicious input could influence the text content returned to the agent.
- Ingestion points: The
search.pyscript takes a user-controlled query argument used to search internal CSV data. - Boundary markers: Absent in the search implementation.
- Capability inventory: The processing logic in
core.pyanddesign_system.pyis restricted to reading local CSV files and outputting formatted text. No network, filesystem-write, or subprocess execution capabilities exist within the runtime scripts. - Sanitization: Input is processed through basic regex-based tokenization in
core.py.
Audit Metadata