skill-authoring

Pass

Audited by Gen Agent Trust Hub on May 12, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: A maintenance script scripts/update-sources.sh is provided to fetch the latest reference documentation and the skills-ref validation library from established domains and GitHub repositories including Anthropic and PyTorch.\n- [COMMAND_EXECUTION]: Local developer tools like scripts/init.py for skill scaffolding and render-graphs.js for diagram generation use system tools such as graphviz and standard filesystem operations to assist in the skill-creation workflow.\n- [SAFE]: The skill provides guidance on security best practices, such as preventing credential exposure in chat logs and avoiding accidental dynamic context execution during documentation, demonstrating a security-first design.
Audit Metadata
Risk Level
SAFE
Analyzed
May 12, 2026, 11:03 PM