The Agent Skills Directory

[SAFE]: No prompt injection or behavior override patterns were detected. The instructions align with providing architectural advice.
[SAFE]: No evidence of data exfiltration, credential harvesting, or sensitive file access was found. The skill does not initiate network connections.
[SAFE]: The skill does not download or execute external code, scripts, or unverifiable dependencies.
[SAFE]: No obfuscation techniques, hidden URLs, or encoded payloads were identified in the skill content or metadata.
[SAFE]: No privilege escalation or persistence mechanisms are present.

design-patterns