The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains instructional constraints to enforce high quality and strict adherence to requirements. These instructions define a professional persona for a reviewer and do not attempt to bypass safety filters or override model constraints.
[DATA_EXFILTRATION]: No sensitive data access, hardcoded credentials, or unauthorized network operations were detected in the instructions or example sub-agent calls.
[REMOTE_CODE_EXECUTION]: The skill uses an internal sub-agent dispatching tool. It does not download or execute code from external or untrusted remote sources.
[COMMAND_EXECUTION]: No dangerous shell commands, privilege escalation attempts, or persistence mechanisms are present.
[SAFE]: The skill implements a standard managerial pattern for agentic workflows with no malicious patterns or security risks identified.

delegate