agent-stop-conditions

Pass

Audited by Gen Agent Trust Hub on Aug 20, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides documentation and implementation patterns for agent safety boundaries, such as retry limits, human approval gates, and resource budgets to prevent unbounded loops and resource exhaustion.
  • [SAFE]: Code snippets are illustrative TypeScript examples intended for development use and do not contain executable malicious scripts or dangerous shell commands.
  • [SAFE]: External links point to informational technical blogs and documentation sites related to AI agent governance and reliability.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 20, 2026, 11:15 PM
Security Audit — agent-trust-hub — agent-stop-conditions