writing-for-agents
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely informational, providing structural and stylistic advice for writing agent instructions. No malicious patterns or behaviors were detected.
- [DYNAMIC_CONTEXT_INJECTION]: The skill explicitly advises against using platform-specific dynamic context injection (e.g., shell command execution at load time), promoting safer and more portable skill design.
- [COMMAND_EXECUTION]: The documentation describes the use of the
allowed-toolsfrontmatter field for pre-approving low-risk commands, aligning with the principle of least privilege.
Audit Metadata