Skill Quality Score Calculator¶
Use the sliders below to score your Agent Skill across seven dimensions. The total score updates in real time along with a letter grade and a verdict. Load a preset to see how a minimal, good, or production-ready skill compares.
Scoring Dimensions¶
| Dimension | Max Points | What It Checks |
|---|---|---|
| Frontmatter Compliance | 20 | Required YAML fields: name, description, version, author, category, tags |
| Structure Compliance | 20 | All required top-level sections present and in order |
| Workflow Quality | 15 | Step clarity, error handling paths, logical flow |
| Quality Scoring System | 15 | Internal success metrics, rubrics, self-evaluation criteria |
| Token Efficiency | 10 | Concise prompts — no bloat, repetition, or over-specification |
| Ecosystem Quality | 10 | Tool declarations, cross-skill integration, discovery metadata |
| Cross-Platform Readiness | 10 | Works in Claude.ai, Claude Code CLI, and API without modification |
Grade Scale¶
| Score | Grade | Verdict |
|---|---|---|
| 85 – 100 | A | Production ready. Ship it. |
| 70 – 84 | B | Solid foundation. Address the gaps before publishing. |
| 55 – 69 | C | Needs work. Focus on the lowest-scoring dimensions. |
| 40 – 54 | D | Significant gaps. Review the guide chapters for each weak area. |
| 0 – 39 | F | Start with Chapter 3: Your First Skill. |