yao-business-skill

Pass

Audited by Gen Agent Trust Hub on Apr 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's primary functionality is driven by a series of Python scripts (assemble_report.py, render_report.py, score_metrics.py, validate_report.py) that perform data validation, scoring, and report assembly. A review of these scripts shows they only use standard Python modules and do not execute shell commands or perform network requests.
  • [SAFE]: Instructions in SKILL.md and associated references focus on structured business analysis workflows. The skill emphasizes distinguishing between facts and hypotheses and requires confidence scores for estimates, which promotes safe and transparent AI behavior.
  • [SAFE]: Sample input and output files reference official business documentation and investor relations pages for well-known services. No high-risk domains, paste sites, or suspicious third-party links are present.
  • [SAFE]: The HTML rendering logic in scripts/render_report.py includes basic sanitization by replacing specific characters (<, >, &) with Unicode escapes before embedding user-derived data into the report template, reducing the risk of content injection in the final report.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 30, 2026, 07:02 PM