Skills
skills/quzhi-ai/logo2docs/quzhi-ai-brand-office-suite/Gen Agent Trust Hub

quzhi-ai-brand-office-suite

Pass

Audited by Gen Agent Trust Hub on May 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill processes user-supplied branding information (logos and style preferences) through deterministic Python scripts. While it ingests external data, the risk of indirect prompt injection is minimized by the structured design-system approach and the use of the brand_check.py utility to validate outputs.
  • [COMMAND_EXECUTION]: The skill executes local Python scripts for core functionality: brand_setup.py for color science, brand_preview.py for HTML generation, and brand_check.py for compliance verification. These scripts use standard libraries (Pillow, openpyxl, etc.) for processing.
  • [EXTERNAL_DOWNLOADS]: The skill manages its own dependencies by checking for required Python packages and automatically installing them via pip from standard registries. It also references its own installation via npx from the author's official repository.
  • [SAFE]: The inclusion of a dedicated compliance checker that specifically scans for local file path leaks in output files demonstrates a proactive security posture to protect user privacy during document generation.
Audit Metadata
Risk Level
SAFE
Analyzed
May 10, 2026, 10:10 AM
Security Audit — agent-trust-hub — quzhi-ai-brand-office-suite