agentic-principles

Pass

Audited by Gen Agent Trust Hub on Apr 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists of documentation and guidelines for AI interaction (e.g., 'Divide and Conquer', 'Context Management'). No executable scripts or automated commands are included.- [SAFE]: References included in the documentation point to official developer guidelines from Anthropic and well-known community repositories related to the author's work. These are used for educational purposes and do not trigger remote code execution.- [SAFE]: The instruction set encourages safety-oriented workflows, such as using 'Plan Mode' for 90% of tasks and performing 'Validation and Retrospectives' after execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 30, 2026, 05:51 AM
Security Audit — agent-trust-hub — agentic-principles