The Agent Skills Directory

[SAFE]: The skill defines a rigorous set of engineering principles aimed at improving software quality and security, specifically instructing the agent to protect contracts, data integrity, and trust boundaries.
[SAFE]: No malicious code, obfuscated URLs, or prompt injection patterns were identified. The instructions explicitly promote defensive programming practices such as input validation and log sanitization.
[SAFE]: The skill includes an 'operating contract' and a 'three-loop limit' mechanism designed to ensure agent transparency and prevent autonomous agents from entering infinite corrective cycles without human intervention.

cleancoding