hermes-agent-skill-authoring
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a meta-instructional guide for developers to author skills within the Hermes ecosystem. It focuses on structure, documentation, and repository standards without introducing executable code or malicious logic.
- [SAFE]: References to sensitive directories (e.g.,
~/.ssh,~/.aws/credentials) and potentially dangerous commands (e.g.,sudo,curl) are provided strictly for educational purposes, teaching authors how to audit for platform-specific dependencies and security risks. - [SAFE]: The content includes explicit security-best-practice guidance, such as warning against using the dynamic context injection (
!command) syntax and advising on the use of least-privilege principles for data processing. - [SAFE]: Mentions of external tools and repositories, including the
ponti-studios/kernellibrary and thenpx skillsmanagement utility, are contextually appropriate for the trusted vendor and intended for legitimate workflow automation.
Audit Metadata