What problem does it solve?
Authoring agent skills without a consistent structure leads to poor routing, vague execution paths, and unmaintainable prompt files. This Skill standardizes how new OMA skills are created, converted, and audited so every SKILL.md is discoverable, executable, and validatable.
Core Features & Use Cases
- SSL-lite Authoring: Generates SKILL.md files with the four required sections (Scheduling, Structural Flow, Logical Operations, References) plus a canonical command or workflow path.
- Skill Conversion & Auditing: Normalizes existing skills into the standard format and checks routing boundaries, resource scope, guardrails, and failure-recovery coverage.
- Validation Workflow: Runs
oma skills lint and a structured checklist covering structure, cross-skill description collisions, and utility content dimensions.
- Use Case: When adding a new capability to your agent harness, ask for a new skill and receive a complete
.agents/skills/{name}/SKILL.md with trigger intents, boundaries, canonical execution path, and a validation report.
Quick Start
Create a new OMA skill named oma-data-export that exports query results to CSV, following the SSL-lite format.