The Agent Skills Directory

[SAFE]: The skill consists entirely of natural language instructions for an AI agent. It does not include any scripts, binaries, or automated tasks. The file serves as a guide for generating high-quality documentation.
[PROMPT_INJECTION]: A heuristic detection for concealment was identified as a false positive. The analyzed content uses the word 'hide' to critique poor instructional design (e.g., 'hide the result until the end') rather than attempting to mask malicious activity, override safety protocols, or disregard agent instructions.

tutorial