Check Catalog Overview

Skill Lab includes 28 static checks organized across 4 evaluation dimensions. Each check validates a specific aspect of skill quality, from basic structural requirements to content optimization.

Check Dimensions

Checks are organized into four dimensions:

Structure - File organization, frontmatter format, folder validation
Naming - Name format and consistency requirements
Content - Body quality, token budgets, examples, and references
Description - Description field requirements

Severity Levels

Each check has a severity level that affects scoring:

Severity	Impact	Description
ERROR	High	Spec violations or critical issues
WARNING	Medium	Quality issues that should be fixed
INFO	Low	Suggestions for improvement

Spec-Required vs. Quality Checks

Of the 28 total checks:

10 checks are spec-required (must pass for spec compliance)
18 checks are quality suggestions (best practices)

Use sklab evaluate -s to run only spec-required checks.

All Checks Summary

Structure Checks (7)

Check ID	Severity	Spec Required	Description
`structure.skill-md-exists`	ERROR	Yes	SKILL.md file exists in the skill directory
`structure.valid-frontmatter`	ERROR	Yes	YAML frontmatter is parseable and valid
`structure.standard-frontmatter-fields`	WARNING	No	Frontmatter contains only fields defined in the Agent Skills spec
`structure.scripts-valid`	WARNING	No	/scripts contains only .py, .sh, .js, .ts, .bash, .rb files
`structure.references-valid`	WARNING	No	/references contains only .md, .txt, .rst files
`structure.scripts-no-interactive`	WARNING	No	Scripts do not use interactive input (agents run non-interactive shells)
`structure.scripts-self-contained`	INFO	No	Scripts folder has no loose dependency manifests

Naming Checks (1)

Check ID	Severity	Spec Required	Description
`naming.matches-directory`	ERROR	Yes	Name must match the parent directory name

Content Checks (11)

Check ID	Severity	Spec Required	Description
`content.body-not-empty`	WARNING	No	SKILL.md body has meaningful content
`content.line-budget`	WARNING	No	Body is under 500 lines
`content.token-budget`	WARNING	No	Body is under 5000 tokens
`content.metadata-token-budget`	INFO	No	Metadata is under 150 tokens for efficient discovery
`content.has-examples`	INFO	No	Content contains code examples
`content.description-actionable`	INFO	No	Description explains when to use this skill
`content.reference-depth`	WARNING	No	References are max 1 level deep
`content.scripts-referenced`	WARNING	No	Scripts in scripts/ are mentioned in the SKILL.md body
`content.script-paths-exist`	WARNING	No	Script paths referenced in body resolve to files on disk
`content.asset-paths-exist`	WARNING	No	Asset paths referenced in body resolve to files on disk
`content.compatibility-prereqs`	INFO	No	Command runners in body are documented in compatibility field

Schema Checks (9)

These checks validate frontmatter fields:

Check ID	Severity	Spec Required	Description
`naming.required`	ERROR	Yes	Name field is present in frontmatter
`naming.format`	ERROR	Yes	Name is lowercase, hyphen-separated, max 64 chars
`description.required`	ERROR	Yes	Description field is present in frontmatter
`description.not-empty`	ERROR	Yes	Description is not empty or whitespace-only
`description.max-length`	ERROR	Yes	Description is under 1024 characters
`frontmatter.compatibility-length`	ERROR	Yes	Compatibility field is under 500 characters if provided
`frontmatter.metadata-format`	ERROR	Yes	Metadata field is a string-to-string mapping if provided
`frontmatter.license-format`	WARNING	No	License field is a string if provided
`frontmatter.allowed-tools-format`	WARNING	No	Allowed-tools field is a space-delimited string if provided

Usage

Run all checks:

sklab evaluate ./my-skill

Run spec-required checks only:

sklab evaluate -s ./my-skill

List all checks:

sklab list-checks

List checks by dimension:

sklab list-checks --dimension structure

Implementation Details

Skill Lab uses two patterns for implementing checks:

Behavioral checks - Hand-written classes with @register_check decorator (structure, naming, content)
Schema checks - Declarative FieldRule definitions that are automatically converted to check classes

Both patterns produce identical CheckResult objects and are executed by the StaticEvaluator.

For detailed information about each check, see the dimension-specific pages:

​Check Dimensions

​Severity Levels

​Spec-Required vs. Quality Checks

​All Checks Summary

​Structure Checks (7)

​Naming Checks (1)

​Content Checks (11)

​Schema Checks (9)

​Usage

​Implementation Details