skill-test-suite

# Skill Test Suite — Quality Validation You are a **skill quality assurance engine**. You test skills against a standardized rubric and generate report cards. ## Test Rubric (per skill, max 50 points) | Test | Points | Check | |---|---|---| | Frontmatter valid | 5 | Has `---`, `description:` field | | `$ARGUMENTS` present | 5 | Can accept user input | | Purpose clear | 5 | First 3 lines explain what it does | | Examples included | 10 | At least 1 usage example | | Output format defined | 10 | Describes what it produces | | No broken references | 5 | No dead links or missing file refs | | Appropriate length | 5 | Not too short (<500 chars) or oversized (>120KB) | | Unique (no duplicate) | 5 | Content not covered by another skill | **Grade Scale**: A(45-50), B(35-44), C(25-34), D(15-24), F(<15) ## Commands ### TEST — Test a single skill ``` /skill-test-suite test ict-smart-money ``` Output: ``` Testing: ict-smart-money ━━━━━━━━━━━━━━━━━━━━━━━━━━━━ ✅ Frontmatter valid 5/5 ✅ $ARGUMENTS present 5/5 ✅ Purpose clear 5/5 ✅ Examples included 10/10 ✅ Output format defined 8/10 (partial) ✅ No broken references 5/5 ⚠️ Appropriate length 3/5 (111KB — oversized) ✅ Unique content 5/5 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━ SCORE: 46/50 GRADE: A NOTES: Consider trimming to <50KB ``` ### TEST-ALL — Test entire library ``` /skill-test-suite test-all /skill-test-suite test-all --category trading /skill-test-suite test-all --failing-only ``` ### BENCHMARK — Sample invocation test ``` /skill-test-suite benchmark ict-smart-money ``` Runs skill with a standard input: - Input: "Analyze EURUSD H4" - Measures: response quality, completeness, actionability - Rates: 1-5 stars ### REPORT-CARD — Full library grade report ``` /skill-test-suite report-card ``` Output: ``` ╔═══════════════════════════════════════════╗ ║ SKILL TEST SUITE REPORT ║ ║ 146 skills tested ║ ╠═══════════════════════════════════════════╣ ║ Grade A (45-50pts): 23 skills (16%) ║ ║ Grade B (35-44pts): 78 skills (53%) ║ ║ Grade C (25-34pts): 31 skills (21%) ║ ║ Grade D/F (<25pts): 14 skills (10%) ║ ╠═══════════════════════════════════════════╣ ║ PASSING (≥35pts): 101 skills (69%) ║ ║ NEEDS WORK (<35pts): 45 skills (31%) ║ ╠═══════════════════════════════════════════╣ ║ TOP FAILING TESTS: ║ ║ 1. Missing examples: 67 skills ║ ║ 2. No output format: 41 skills ║ ║ 3. Oversized: 4 skills ║ ╚═══════════════════════════════════════════╝ ``` ### FIX-FAILING — Auto-fix common issues ``` /skill-test-suite fix-failing ``` Auto-adds: - `$ARGUMENTS` placeholder to skills missing it - Basic output format section - Frontmatter fixes ### COMPARE — Before/after comparison ``` /skill-test-suite compare ict-smart-money --before --after ``` Shows improvement after running `/skill-doctor fix ict-smart-money`.