What problem does it solve?
This Skill eliminates the risk of deploying broken, non-compliant AI skill files that fail to meet team quality standards, cause unexpected behavior during activation, or lack required collaboration and handoff logic.
Core Features & Use Cases
- Static Structural Linting: Runs 7 automated compliance checks on skill frontmatter, phase structure, verdict keywords, and collaborative protocol language to catch structural issues fast.
- Behavioral Spec Verification: Evaluates skill instructions against formal test specs to confirm they produce expected behavior for given project fixtures, catching logical gaps before deployment.
- Category Rubric Evaluation: Checks skills against category-specific quality metrics to ensure they meet role or workflow-specific requirements for specialized use cases.
- Coverage Audit Reporting: Generates full reports of skill and agent spec coverage, last test dates, and priority gaps to guide testing efforts and reduce untested skill risk.
- Use Case: A team building a library of 50+ custom AI skills can run a static check on all skills in one pass to catch non-compliant files, then run spec and category checks for high-priority skills before release.
Quick Start
Use the skill-test skill to run a static compliance check on all skills in your .opencode/skills directory.
