authoring-tips

Pass

Audited by Gen Agent Trust Hub on May 4, 2026

Risk Level: SAFE
Full Analysis
  • [Instructional Directives]: The file prevent-child-agent-responses.md includes strong instructional markers like 'CRITICAL
  • DO NOT MESSAGE USERS'. While these are intended as templates for agent authors to apply to child agents, they are presented as documentation snippets rather than attempts to override the current agent's core safety or behavior guidelines.
  • [Dynamic Content Integration]: The skill documents the use of Power Fx expressions (e.g., {Text(Today(), ...)}) for providing temporal context in date-context.md. These are standard platform-specific templating features and do not involve unauthorized shell access or external command execution.
  • [Runtime Logic Patterns]: The Topic-redirect-withvariable.md file describes a method for dynamic topic routing using Switch() expressions. This pattern facilitates internal agent navigation within the platform's schema and does not represent a security risk.
Audit Metadata
Risk Level
SAFE
Analyzed
May 4, 2026, 10:53 PM