qt6-desktop-ux
Pass
Audited by Gen Agent Trust Hub on Feb 23, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill directs the agent to retrieve external data via documentation lookup tools, creating a surface for potential tool output poisoning. 1. Ingestion points: mcp__context7__get-library-docs in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: Subprocess calls (python) and MCP tool execution. 4. Sanitization: Absent.
- [Command Execution] (SAFE): Includes usage templates for a local theme management utility (scripts/add_theme_color.py), which is a standard developer automation pattern.
Audit Metadata