Guides creation, validation, and packaging of AI agent skills with token-efficient design, progressive disclosure patterns, and YAML frontmatter best practices. Use when building new skills, updating existing skills, validating skill structure against standards, or packaging for distribution—e.g., "create skill", "validate SKILL.md", "package skill for sharing", "check description format".
93
92%
Does it follow best practices?
Impact
93%
1.75xAverage score across 3 eval scenarios
Passed
No known issues
Correct skill structure and description format
Kebab-case name
100%
100%
Frontmatter fields only
0%
100%
Description capability statement
100%
100%
Description Use when clause
0%
100%
Description examples present
0%
100%
Description word count
0%
100%
Purpose section
100%
100%
No auxiliary docs
100%
100%
Imperative form
100%
100%
Line count
100%
100%
References linked with guidance
42%
100%
No Use when in body
20%
100%
Script execution workflow (shebang vs python)
init_skill shebang invocation
0%
100%
quick_validate shebang invocation
0%
100%
package_skill shebang invocation
0%
100%
chmod executable
0%
0%
validate before package
100%
100%
.skill file produced
100%
100%
SKILL.md Purpose section
100%
100%
No auxiliary docs
100%
100%
Description has Use when
100%
100%
Persuasion principles for discipline-enforcing skill
Authority language present
100%
100%
Commitment mechanism
53%
100%
Social proof language
80%
100%
No Liking principle
100%
100%
No fear or shame tactics
100%
100%
Ethical note included
37%
37%
Use when in frontmatter only
0%
100%
Purpose section
0%
100%
Imperative form
100%
100%
Kebab-case name
0%
100%
b74de5e
Table of Contents
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.