The Agent Skills Directory

[SAFE]: The skill provides a detailed, methodology-driven framework for debugging (application debugging, flaky tests, race conditions). It defines clear phases: reproduction, hypothesis formation, observation, and verification.
[SAFE]: All commands mentioned (such as omh runtime record or shell loops for test repetition) are standard development practices for debugging and do not involve unauthorized access, sensitive data exfiltration, or malicious execution.
[SAFE]: The skill explicitly includes safety rules regarding data handling, such as redacting credentials and secrets from output logs before recording them (redact before recording).
[SAFE]: It utilizes standard external references for debugging methods (e.g., systematic-debugging) which are identified as bundled or official optional components.
[SAFE]: No obfuscation, prompt injection, or persistence mechanisms were detected. The use of non-English terms in the 'Strong routing signals' (e.g., 'почини баг') is legitimate localization for a debugging tool and not a homoglyph attack.

omh-app-debugging