observability-first
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of documentation and guidelines for diagnostic methodology and software development. It does not include any scripts, tools, or executable commands.
- [DATA_EXPOSURE]: The skill promotes security best practices by explicitly instructing the agent to never log secrets, credentials, or sensitive payloads (e.g., 'Never log secrets or huge payloads by default', 'What not to emit: ... anything from secret-hygiene').
- [INDIRECT_PROMPT_INJECTION]: While the skill encourages the use of external logs for debugging, it provides conceptual guidance rather than automated parsing logic, and its focus on structured summaries and minimal logging reduces the surface area for processing untrusted data.
Audit Metadata