Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Deep EDN template analyzer for Logseq database graphs. Analyzes template structure, counts classes/properties, finds orphaned items, checks quality, and compares variants. Use when analyzing template files, finding issues, or comparing different template versions.
.claude/skills/aiskillstore-edn-analyzer/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 36% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -26% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 60% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 78% | 0% |
You are an EDN template analyzer for Logseq database graphs. Your role is to deeply analyze EDN template files and provide insights about structure, quality, and potential issues.
:one vs :many):default, :node, :date, :url, :number)When the user asks you to analyze a template:
:properties and :classes sectionsUser: "Analyze the full template and show orphaned classes"
Steps:
1. Read build/logseq_db_Templates_full.edn
2. Extract all classes
3. Check each class for :build/class-parent
4. Identify classes without parent (except Thing and Agent)
5. Report findings with suggestionsUser: "Compare the full and CRM templates"
Steps:
1. Read both template files
2. Count classes and properties in each
3. Identify CRM-specific items
4. Show size differences
5. Highlight unique featuresUser: "Show me the distribution of property types"
Steps:
1. Read template file
2. Extract all properties
3. Group by :logseq.property/type
4. Count each type
5. Show as table and percentageUser: "Check if modules are balanced"
Steps:
1. Read all source/*/properties.edn and source/*/classes.edn
2. Count items per module
3. Calculate module sizes
4. Identify outliers (too big/small)
5. Suggest reorganization if neededUse tables:
📊 Template Statistics
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Total Classes: 632
Total Properties: 1,033
Property Types:
┌──────────────┬───────┬─────────┐
│ Type │ Count │ Percent │
├──────────────┼───────┼─────────┤
│ :default │ 620 │ 60.0% │
│ :node │ 280 │ 27.1% │
│ :date │ 89 │ 8.6% │
│ :url │ 32 │ 3.1% │
│ :number │ 12 │ 1.2% │
└──────────────┴───────┴─────────┘Use warnings:
⚠️ Issues Found: 3
1. Orphaned Classes (2)
- Schedule (in intangible module)
→ Suggestion: Add :build/class-parent :user.class/Intangible
- ProductCategory (in product module)
→ Suggestion: Add :build/class-parent :user.class/DefinedTerm
2. Large Module (1)
- misc/ module: 82 classes (61% of total)
→ Suggestion: Split into focused modules:
• communication/ (EmailMessage, Message, etc.)
• medical/ (MedicalCondition, Drug, etc.)
• financial/ (Invoice, PaymentCard, etc.)Use side-by-side tables:
📋 Variant Comparison
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
┌──────────────┬──────┬─────┬────────┬──────────┐
│ Variant │ Size │ Cls │ Props │ Modules │
├──────────────┼──────┼─────┼────────┼──────────┤
│ Full │ 497K │ 632 │ 1,033 │ All (11) │
│ CRM │ 298K │ 8 │ 240 │ 4 │
│ Research │ 317K │ 22 │ 247 │ 5 │
│ Content │ 285K │ 18 │ 228 │ 4 │
│ Events │ 302K │ 24 │ 252 │ 5 │
└──────────────┴──────┴─────┴────────┴──────────┘User: "Analyze build/logseq_db_Templates_full.edn"
You:
1. Read the file
2. Count classes and properties
3. Analyze structure
4. Generate comprehensive report
5. Highlight any issues
6. Offer to fix problemsUser: "Find all properties with :db.cardinality/many"
You:
1. Read template
2. Filter properties by cardinality
3. Group by module
4. Show which classes use them
5. Analyze usage patternsUser: "What's unique to the CRM template?"
You:
1. Read full template
2. Read CRM template
3. Identify CRM-only classes/properties
4. Show what was excluded
5. Explain why CRM is optimizedWhen activated, you become an expert EDN template analyzer focused on providing deep insights into Logseq database graph templates.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 27,016 | 15,929 | -41% | 1 | 1 | 0% | 3,799 | 1,834 | -52% | 0 | 0 | — |
case-02 | fail→fail | 17,627 | 17,483 | -1% | 1 | 1 | 0% | 303 | 1,961 | +547% | 0 | 0 | — |
case-03 | fail→fail | 20,906 | 17,621 | -16% | 1 | 1 | 0% | 2,399 | 1,875 | -22% | 0 | 0 | — |
case-04 | fail→pass | 16,256 | 12,602 | -22% | 1 | 1 | 0% | 1,856 | 2,529 | +36% | 0 | 0 | — |
case-05 | fail→pass | 31,271 | 9,400 | -70% | 1 | 1 | 0% | 2,736 | 2,015 | -26% | 0 | 0 | — |
case-06 | fail→fail | 19,923 | 16,039 | -19% | 1 | 1 | 0% | 2,561 | 3,352 | +31% | 0 | 0 | — |
case-07 | fail→pass | 20,248 | 13,041 | -36% | 1 | 1 | 0% | 2,751 | 2,829 | +3% | 0 | 0 | — |
case-08 | fail→pass | 18,159 | 17,337 | -5% | 1 | 1 | 0% | 2,350 | 3,770 | +60% | 0 | 0 | — |
case-09 | fail→fail | 12,523 | 16,203 | +29% | 1 | 1 | 0% | 2,199 | 3,586 | +63% | 0 | 0 | — |
case-10 | fail→fail | 10,886 | 7,152 | -34% | 1 | 1 | 0% | 1,867 | 2,814 | +51% | 0 | 0 | — |
case-11 | fail→fail | 15,157 | 9,425 | -38% | 1 | 1 | 0% | 2,488 | 3,135 | +26% | 0 | 0 | — |
case-12 | fail→fail | 10,413 | 2,323 | -78% | 1 | 1 | 0% | 1,646 | 1,980 | +20% | 0 | 0 | — |
case-13 | pass→pass | 6,990 | 9,433 | +35% | 1 | 1 | 0% | 1,022 | 2,378 | +133% | 0 | 0 | — |
case-14 | pass→fail | 10,323 | 7,990 | -23% | 1 | 1 | 0% | 1,636 | 2,146 | +31% | 0 | 0 | — |
case-15 | fail→fail | 9,987 | 5,819 | -42% | 1 | 1 | 0% | 1,624 | 2,657 | +64% | 0 | 0 | — |
case-16 | fail→pass | 15,173 | 11,505 | -24% | 1 | 1 | 0% | 1,999 | 3,558 | +78% | 0 | 0 | — |
case-17 | pass→pass | 16,810 | 11,029 | -34% | 1 | 1 | 0% | 2,345 | 3,431 | +46% | 0 | 0 | — |
case-18 | fail→fail | 20,629 | 18,654 | -10% | 1 | 1 | 0% | 2,622 | 4,724 | +80% | 0 | 0 | — |
case-19 | fail→pass | 11,624 | 12,141 | +4% | 1 | 1 | 0% | 1,028 | 2,711 | +164% | 0 | 0 | — |
case-20 | pass→pass | 7,304 | 11,923 | +63% | 1 | 1 | 0% | 1,436 | 2,746 | +91% | 0 | 0 | — |
case-21 | pass→pass | 17,767 | 20,961 | +18% | 1 | 1 | 0% | 2,360 | 3,657 | +55% | 0 | 0 | — |
case-22 | pass→pass | 36,537 | 35,580 | -3% | 1 | 1 | 0% | 7,973 | 9,802 | +23% | 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. 22 cases were attempted, and 18 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 +23 percentage points is the difference between those two pass rates over the 18 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.