vitest

I took a look at your vitest skill and wanted to share some thoughts.

Links:

• GitHub
• Your agentic skill in the SkillzWave marketplace

The TL;DR

You're at 75/100, solid C-grade territory. This is based on Anthropic's skill best practices rubric. Your strongest area is Spec Compliance (14/15) — the fundamentals are tight. Weakest is Progressive Disclosure Architecture (22/30) — there's some structural inefficiency that's eating your token economy.

What's Working Well

• Spec compliance is basically perfect (14/15). Your YAML frontmatter is clean, naming follows conventions, and your description has comprehensive trigger coverage (40+ patterns). This skill will activate on natural queries.
• Your trigger list is genuinely thorough — covering everything from explicit vi.fn, .test.ts, describe( to implicit "mock a service test" patterns. That's good UX design.
• The quick_start table is a solid entry point with clear scenarios, and your success_criteria checklist gives devs a real validation path. That's the kind of thing that makes a skill feel approachable.

The Big One: Token Waste From Duplication

Your mocking patterns appear twice — once inline in the <mocking> section (lines 110-123) and again in templates/mock-module.ts.md. For a 350-line file, this is burning tokens on repetition.

The fix: Keep only the vi.fn() basics in SKILL.md with a reference to the template. Move the full module mocking example into the template exclusively. This alone should save ~40 tokens and bump you +2 points toward 77/100.

Other Things Worth Fixing

1. Add a table of contents — 350+ lines without a TOC makes navigation a pain. Add a simple ## Contents section listing: objective, mocking, advanced patterns, assertions, templates, commands. Gets you +2 points.

2. Use references/ instead of templates/ — that's the standard directory name in the skill spec. Small change, but standardization matters for discoverability. +1 point.

3. Templates need "Expected Output" sections — your placeholder tables show what to fill in, but not what a passing vitest run actually looks like. Add sample terminal output so devs know what success looks like. +1 point.

4. Missing step-by-step workflow — you have a quick_start table, but no numbered "Step 1... Step 2..." flow for writing a complete test file from scratch. Add a workflow section. +1 point.

Quick Wins

• Deduplicate mocking examples (SKILL.md → template) — +2 pts
• Add table of contents — +2 pts
• Rename templates/ to references/ — +1 pt
• Add expected output to templates — +1 pt

That's 6 points within reach. You'd hit 81/100 and move into B-grade territory with focused work on PDA structure.

Checkout your skill here: SkillzWave.ai | SpillWave We have an agentic skill installer that install skills in 14+ coding agent platforms. Check out this guide on how to improve your agentic skills.