The Agent Skills Directory

[SAFE]: The skill provides operational instructions to ensure rigorous verification of work through standard developer tooling. It does not contain any code, network requests, or attempts to bypass security controls.
[SAFE]: The instructions use strong language (e.g., "Iron Law", "Omerta Compliance") to enforce agent behavior, but these are instructional constraints rather than prompt injections designed to override safety filters or system prompts.
[SAFE]: No obfuscation, data exfiltration, or persistence mechanisms were detected in the skill content.

sweep-verification