The Agent Skills Directory

[SAFE]: The skill provides instructions for self-correction and error recovery, which are positive safety behaviors intended to prevent hallucinations and unproductive tool usage.
[SAFE]: The protocol includes standard developer workflows, such as using git status to verify local changes and reverting files if they are broken, which pose no security risk.
[SAFE]: There are no detected patterns of prompt injection, data exfiltration, obfuscation, or remote code execution. The skill operates entirely within the local context to re-orient the agent to system constraints and user requests.

emergency-confusion-reset