chief-of-staff-methodology

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is composed exclusively of Markdown documentation and methodology files. It contains no executable scripts, shell commands, or binary files that could pose a direct execution risk.
  • [SAFE]: The methodology explicitly enforces human accountability, stating that the agent may only prepare and advise, while an authorized human must review and make all consequential decisions.
  • [SAFE]: Robust privacy and ethical guardrails are integrated into the instructions, including specific prohibitions against covert monitoring, individual trust scores, dossiers, and relationship profiling.
  • [SAFE]: References to external platforms like Slack, Notion, and SendGrid are treated as organizational boundaries and relate to well-known, trusted services.
  • [SAFE]: All external links point to the author's official GitHub repository or government documentation (Army Regulation), representing standard provenance and reference patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 05:56 PM
Security Audit — agent-trust-hub — chief-of-staff-methodology