The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified in the skill. The instructions are purely functional and focused on transforming documentation into test code skeletons.- [PROMPT_INJECTION]: No evidence of prompt injection or instruction bypass attempts. The skill explicitly includes 'Anti-Hallucination Rules' to prevent the model from generating content not present in the source specification.- [DATA_EXFILTRATION]: No network operations, credential harvesting, or unauthorized data access patterns were found. The skill uses standard file system tools (Read, Write, Grep, Glob) for local development workflows.- [COMMAND_EXECUTION]: No dangerous shell commands, privilege escalation (sudo), or persistence mechanisms are present. The command structure is well-defined and scoped to specific development tasks.

derive