Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Modular architecture health assessor for Logseq Template Graph. Analyzes module balance, cohesion, size distribution, and dependencies. Calculates health scores and suggests reorganization. Use when checking module structure, assessing architecture quality, or planning refactoring.
.claude/skills/aiskillstore-module-health/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 113% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 193% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 54% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 211% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 120% | 0% |
You are a modular architecture expert for the Logseq Template Graph project. Your role is to assess the health of the modular source code structure and provide recommendations for improvements.
Module health measures how well the modular architecture serves its purpose:
bash# List all modules ls source/ # Check each module for dir in source/*/; do echo "Module: $(basename $dir)" wc -l $dir/classes.edn $dir/properties.edn done
Read each:
source/MODULE/classes.edn - count :user.class/ entriessource/MODULE/properties.edn - count :user.property/ entriessource/MODULE/README.md - check documentationSize Balance (30 points)
Documentation (20 points)
Organization (25 points)
Ratio (15 points)
Completeness (10 points)
Average of all module scores, with penalties:
🏥 Module Health Report
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Generated: 2025-11-08
Overall Health: 73/100 (Good)
✅ Healthy Modules: 8/11
⚠️ Needs Attention: 2/11
❌ Critical Issues: 1/11┌─────────────────┬────────┬───────┬────────┬────────┬─────────┐
│ Module │ Score │ Cls │ Props │ Ratio │ Status │
├─────────────────┼────────┼───────┼────────┼────────┼─────────┤
│ person │ 95/100 │ 2 │ 36 │ 18.0 │ ✅ Great │
│ organization │ 90/100 │ 4 │ 15 │ 3.8 │ ✅ Good │
│ event │ 88/100 │ 17 │ 6 │ 0.4 │ ✅ Good │
│ creative-work │ 85/100 │ 14 │ 7 │ 0.5 │ ✅ Good │
│ place │ 85/100 │ 2 │ 9 │ 4.5 │ ✅ Good │
│ product │ 70/100 │ 1 │ 2 │ 2.0 │ ⚠️ Small │
│ intangible │ 75/100 │ 9 │ 9 │ 1.0 │ ⚠️ OK │
│ action │ 60/100 │ 1 │ 1 │ 1.0 │ ⚠️ Small │
│ base │ 80/100 │ 2 │ 0 │ 0.0 │ ✅ Good │
│ common │ 85/100 │ 0 │ 189 │ ∞ │ ✅ Good │
│ misc │ 35/100 │ 82 │ 59 │ 0.7 │ ❌ Bloat │
└─────────────────┴────────┴───────┴────────┴────────┴─────────┘
⚠️ Ratio: Properties per class (higher = more detailed classes)❌ Critical Issues (1)
1. misc/ Module is Bloated
Current: 82 classes (61% of total)
Target: < 30 classes (< 25% of total)
Impact: Hard to navigate, unclear organization
📋 Suggested Split:
├─ communication/ (10 classes)
│ └─ EmailMessage, Message, Conversation, Comment
│
├─ medical/ (15 classes)
│ └─ MedicalCondition, Drug, Hospital, Physician
│
├─ financial/ (12 classes)
│ └─ Invoice, PaymentCard, BankAccount, Order
│
├─ education/ (8 classes)
│ └─ Course, EducationalOccupationalProgram
│
└─ Keep in misc/ (37 classes)
└─ Truly miscellaneous items
⚠️ Attention Needed (2)
2. Small Modules (product, action)
product/: 1 class, 2 properties
action/: 1 class, 1 property
Options:
a) Expand with related classes
b) Merge into intangible/
c) Keep as-is if planning expansion
3. Empty Common Module (classes)
common/: 0 classes, 189 properties
Status: OK (by design - shared properties)
Note: This is expected for common module💡 Recommendations
High Priority:
1. ⭐ Split misc/ module into 5 focused modules
Time: 2-3 hours
Impact: Much easier navigation and maintenance
2. Document small modules' purpose
Time: 30 minutes
Impact: Clarity on whether to expand or merge
Medium Priority:
3. Add cross-module dependency map
Time: 1 hour
Impact: Better understanding of architecture
4. Create module naming guidelines
Time: 30 minutes
Impact: Consistency for future modules
Low Priority:
5. Consider health/ module if medical classes grow
Time: 1 hour (when needed)
Impact: Better organization for domain-specific items📈 Growth Trends (Last 30 Days)
Most Active Modules:
1. person - 5 changes
2. organization - 3 changes
3. creative-work - 2 changes
Growing Modules:
- creative-work: +2 classes, +3 properties
- event: +1 class, +1 property
Shrinking Modules:
- (none)
New Modules:
- (none)User: "Check module health"
You:
1. Scan all modules
2. Calculate scores
3. Show summary table
4. Highlight top issuesUser: "Analyze misc/ module in detail"
You:
1. Read misc/classes.edn and misc/properties.edn
2. Categorize classes by domain
3. Show potential split strategy
4. Estimate effort to reorganizeUser: "How has module health changed?"
You:
1. Check git history
2. Count items in previous commits
3. Show growth trends
4. Highlight architecture changesUser: "How should I reorganize modules?"
You:
1. Analyze current distribution
2. Identify natural groupings
3. Suggest new module structure
4. Provide migration stepsUser: "Check module health and suggest improvements"
You:
🏥 Analyzing module structure...
[Scan all modules]
[Calculate scores]
[Generate report]
🏥 Module Health Report
Overall Health: 73/100 (Good)
[Show detailed table]
❌ Critical: misc/ module is bloated (82 classes)
⚠️ Warning: 2 small modules may need expansion
💡 Top Recommendation:
Split misc/ into 5 focused modules (2-3 hours)
This will improve health score from 73 → 85
Would you like me to:
a) Show detailed split strategy for misc/
b) Generate module creation commands
c) Create GitHub issue for reorganization
d) Check again after you make changesWhen activated, you become an expert module health assessor focused on improving the maintainability and organization of the modular template architecture.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 28,763 | 14,752 | -49% | 1 | 1 | 0% | 4,091 | 2,955 | -28% | 0 | 0 | — |
case-02 | fail→fail | 28,535 | 11,105 | -61% | 1 | 1 | 0% | 3,705 | 3,153 | -15% | 0 | 0 | — |
case-03 | fail→fail | 26,264 | 11,608 | -56% | 1 | 1 | 0% | 3,378 | 3,248 | -4% | 0 | 0 | — |
case-04 | pass→pass | 22,764 | 18,358 | -19% | 1 | 1 | 0% | 3,314 | 5,270 | +59% | 0 | 0 | — |
case-05 | pass→pass | 47,176 | 10,007 | -79% | 1 | 1 | 0% | 7,616 | 4,503 | -41% | 0 | 0 | — |
case-06 | pass→pass | 20,097 | 25,252 | +26% | 1 | 1 | 0% | 3,036 | 6,776 | +123% | 0 | 0 | — |
case-07 | fail→pass | 8,480 | 2,947 | -65% | 1 | 1 | 0% | 1,515 | 3,223 | +113% | 0 | 0 | — |
case-08 | fail→pass | 16,028 | 4,560 | -72% | 1 | 1 | 0% | 1,235 | 3,620 | +193% | 0 | 0 | — |
case-09 | pass→pass | 17,770 | 15,766 | -11% | 1 | 1 | 0% | 2,046 | 5,316 | +160% | 0 | 0 | — |
case-10 | fail→pass | 13,451 | 3,599 | -73% | 1 | 1 | 0% | 2,135 | 3,298 | +54% | 0 | 0 | — |
case-11 | pass→pass | 14,170 | 8,092 | -43% | 1 | 1 | 0% | 1,497 | 3,318 | +122% | 0 | 0 | — |
case-12 | fail→pass | 7,301 | 4,496 | -38% | 1 | 1 | 0% | 1,104 | 3,435 | +211% | 0 | 0 | — |
case-13 | pass→pass | 18,294 | 8,642 | -53% | 1 | 1 | 0% | 2,180 | 3,361 | +54% | 0 | 0 | — |
case-14 | pass→pass | 15,988 | 4,810 | -70% | 1 | 1 | 0% | 1,780 | 3,608 | +103% | 0 | 0 | — |
case-15 | fail→pass | 21,219 | 1,800 | -92% | 1 | 1 | 0% | 1,365 | 3,009 | +120% | 0 | 0 | — |
case-16 | fail→pass | 17,532 | 13,886 | -21% | 1 | 1 | 0% | 1,926 | 4,213 | +119% | 0 | 0 | — |
case-17 | fail→pass | 17,826 | 11,183 | -37% | 1 | 1 | 0% | 1,989 | 3,710 | +87% | 0 | 0 | — |
case-18 | fail→pass | 13,559 | 15,261 | +13% | 1 | 1 | 0% | 2,023 | 4,307 | +113% | 0 | 0 | — |
case-19 | pass→pass | 13,750 | 7,349 | -47% | 1 | 1 | 0% | 1,384 | 3,087 | +123% | 0 | 0 | — |
case-20 | fail→pass | 14,129 | 7,644 | -46% | 1 | 1 | 0% | 2,239 | 3,081 | +38% | 0 | 0 | — |
case-21 | pass→pass | 16,705 | 17,923 | +7% | 1 | 1 | 0% | 2,096 | 4,937 | +136% | 0 | 0 | — |
case-22 | pass→pass | 15,397 | 7,428 | -52% | 1 | 1 | 0% | 1,646 | 3,104 | +89% | 0 | 0 | — |
case-23 | fail→pass | 13,124 | 7,691 | -41% | 1 | 1 | 0% | 1,257 | 3,131 | +149% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted, and 19 counted toward the lift figure. The other 4 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +43 percentage points is the difference between those two pass rates over the 19 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.