skill-best-practices
Pass
Audited by Gen Agent Trust Hub on May 3, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a local validation script (scripts/validate-metadata.py) and perform standard file system operations like creating directory structures as part of its core authoring workflow. These operations are restricted to the local environment and are necessary for the skill's intended purpose.
- [SAFE]: The included Python validation script (scripts/validate-metadata.py) uses standard library modules (re, sys, argparse) to perform basic string validation and regex matching. It does not perform network operations, access sensitive files, or utilize dangerous functions such as eval() or exec().
- [SAFE]: The skill body, templates, and checklists are designed for transparency and deterministic behavior. There is no evidence of prompt injection, multi-layer obfuscation, or data exfiltration attempts. The skill adheres to the progressive disclosure and imperative style principles it advocates.
Audit Metadata