The Agent Skills Directory

[SAFE]: The skill provides instructions for a specific persona and interaction style (constructive interrogation). It does not contain any malicious patterns such as prompt injection, data exfiltration, or unauthorized command execution.
[SAFE]: While the skill instructs the agent to inspect the local codebase, documents, and configurations, this is the intended functionality for a design review tool and occurs within the existing context provided to the agent by the user.
[SAFE]: No obfuscation, external dependencies, or persistence mechanisms were detected in the provided content.

grill-me