cross-harness-skills

Pass

Audited by Gen Agent Trust Hub on Oct 5, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
  • [SAFE]: The skill consists entirely of documentation and best practices for cross-platform skill development. It does not include any scripts, commands, or executable code.
  • [SAFE]: Instructions emphasize safety, including requirements for explicit authority for destructive actions and avoiding the exposure of secrets.
  • [EXTERNAL_DOWNLOADS]: The skill references official developer documentation from well-known services (OpenAI and Anthropic) for prompt engineering guidance.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 5, 2026, 11:36 PM