The Agent Skills Directory

[SAFE]: The skill serves as a technical reference guide for writing and structuring AI agent skills effectively. It provides theoretical frameworks for improving the predictability of agent behavior.
[SAFE]: The content consists purely of documentation in Thai and English, covering concepts such as invocation types (model-invoked vs user-invoked), information hierarchy, and common failure modes in prompt engineering.
[SAFE]: No executable scripts, command-line operations, or network requests are defined in the skill. It utilizes only markdown text and standard relative file links between SKILL.md and GLOSSARY.md.
[SAFE]: The disable-model-invocation: true frontmatter configuration is appropriately used to ensure the skill acts as a manual reference for the user or other skills rather than an autonomous tool.
[SAFE]: No obfuscation, data exfiltration patterns, or suspicious external dependencies were found during the analysis.

writing-great-skills