[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6EMTGS2QZKE27CKBNZPJCQ3":3,"skill-same:01M6EMTGS2QZKE27CKBNZPJCQ3":62},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":28,"linkOnly":29,"markdownContent":30,"markdownBody":31,"frontMatter":32,"refName":33,"commitSha":34,"skillDirectory":35,"forksCount":36,"repositoryHtmlUrl":37,"repositoryHomepageUrl":38,"ownerHtmlUrl":39,"ownerBlogUrl":6,"ownerDescription":40,"updatedAt":41,"install":42,"resources":55},"01M6EMTGS2QZKE27CKBNZPJCQ3","reasoning-inductive",null,"Extract patterns and generalizations from multiple observations. Use when detecting recurring themes, building predictive rules, or identifying systemic behaviors from accumulated data. Produces validated patterns with confidence bounds and exception handling.","ComeOnOliver\u002Fskillshub","ComeOnOliver","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F43215937?v=4","01M6E5JNFECJAAVFE0W7SMWWD2","Haoran Shu","01M6E5JNNDQS7AJZDYFQ8TPZCK",66,"MIT","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Faiskillstore\u002Fmarketplace\u002Fbellabe\u002Freasoning-inductive\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27],"reasoning","reasoning inductive","data","Rules","extract","recurring","use","patterns","themes","2026-11-10 10:12:30",false,"---\nname: reasoning-inductive\ndescription: Extract patterns and generalizations from multiple observations. Use when detecting recurring themes, building predictive rules, or identifying systemic behaviors from accumulated data. Produces validated patterns with confidence bounds and exception handling.\n---\n\n# Inductive Reasoning\n\nGeneralize from instances to rules. The logic of pattern extraction and empirical learning.\n\n## Type Signature\n\n```\nInductive : [Observation] → Pattern → Generalization → ConfidenceBounds\n\nWhere:\n  Observations     : [Instance] → Dataset\n  Pattern          : Dataset → (Regularity × Frequency)\n  Generalization   : (Regularity × Frequency) → Rule\n  ConfidenceBounds : Rule × SampleSize → (Confidence × Exceptions)\n```\n\n## When to Use\n\n**Use inductive when:**\n- Multiple similar observations accumulate\n- Looking for recurring patterns across threads\n- Building predictive rules from experience\n- Identifying systemic behaviors\n- Validating or discovering Canvas assumptions\n- \"This keeps happening\" situations\n\n**Don't use when:**\n- Explaining single observation → Use Abductive\n- Known causal chain exists → Use Causal\n- Transferring one case to another → Use Analogical\n- Resolving disagreement → Use Dialectical\n\n## Distinction from Other Modes\n\n| Mode | Input | Output | Question |\n|------|-------|--------|----------|\n| **Abductive** | Single anomaly | Explanation | \"Why did this happen?\" |\n| **Inductive** | Multiple instances | Pattern\u002FRule | \"What keeps happening?\" |\n| **Analogical** | One source case | Transferred solution | \"How is this like that?\" |\n\n**Key difference from Abductive:**\n- Abductive: 1 observation → 1 explanation\n- Inductive: N observations → 1 generalization\n\n## Four-Stage Process\n\n### Stage 1: Observation Collection\n\n**Purpose:** Gather and structure multiple instances for analysis.\n\n**Minimum Sample Requirements:**\n\n| Confidence Target | Minimum N | Notes |\n|-------------------|-----------|-------|\n| Exploratory | 3-5 | Hypothesis generation only |\n| Tentative | 6-10 | Directional confidence |\n| Moderate | 11-20 | Actionable patterns |\n| High | 21+ | Strong generalizations |\n\n**Components:**\n```yaml\nobservations:\n  dataset:\n    - instance_id: \"deal-001\"\n      timestamp: ISO8601\n      context: \"Enterprise sales\"\n      attributes:\n        deal_size: 400000\n        sales_cycle: 120\n        stalled_at: \"legal_review\"\n        outcome: \"won\"\n        \n    - instance_id: \"deal-002\"\n      timestamp: ISO8601\n      context: \"Enterprise sales\"\n      attributes:\n        deal_size: 350000\n        sales_cycle: 150\n        stalled_at: \"legal_review\"\n        outcome: \"lost\"\n        \n    # ... more instances\n    \n  metadata:\n    total_instances: 12\n    time_range: \"Q3-Q4 2024\"\n    source: \"threads\u002Fsales\u002F*\u002F6-learning.md\"\n    collection_method: \"automated scan\"\n    \n  quality:\n    completeness: 0.92  # % of fields populated\n    consistency: 0.88   # % following same schema\n    recency: 0.75       # Weight toward recent\n```\n\n### Stage 2: Pattern Detection\n\n**Purpose:** Identify regularities in the dataset.\n\n**Pattern Types:**\n\n| Type | Description | Example |\n|------|-------------|---------|\n| **Frequency** | How often X occurs | \"7\u002F12 deals stall at legal\" |\n| **Correlation** | X and Y co-occur | \"Large deals AND long cycles\" |\n| **Sequence** | X follows Y | \"Stall → lose within 30 days\" |\n| **Cluster** | Groups emerge | \"Two deal archetypes exist\" |\n| **Trend** | Direction over time | \"Cycles getting longer\" |\n| **Threshold** | Breakpoint exists | \"Deals >$300K behave differently\" |\n\n**Detection Process:**\n```yaml\npatterns:\n  detected:\n    - pattern_id: P1\n      type: frequency\n      description: \"Legal review stalls\"\n      evidence: \"7 of 12 deals (58%) stalled at legal review\"\n      strength: 0.78\n      \n    - pattern_id: P2\n      type: correlation\n      description: \"Deal size correlates with cycle length\"\n      evidence: \"r=0.72 between deal_size and sales_cycle\"\n      strength: 0.72\n      \n    - pattern_id: P3\n      type: threshold\n      description: \"CFO involvement threshold\"\n      evidence: \"Deals >$250K require CFO, adding 30+ days\"\n      strength: 0.85\n      \n    - pattern_id: P4\n      type: sequence\n      description: \"Stall duration predicts outcome\"\n      evidence: \"Stalls >21 days → 80% loss rate\"\n      strength: 0.80\n      \n  rejected:\n    - pattern: \"Industry affects outcome\"\n      reason: \"No significant difference across industries (p>0.3)\"\n      \n  insufficient_data:\n    - pattern: \"Seasonality effects\"\n      reason: \"Only 2 quarters of data, need 4+ for seasonality\"\n```\n\n### Stage 3: Generalization\n\n**Purpose:** Form rules from validated patterns.\n\n**Rule Formation:**\n```yaml\ngeneralizations:\n  rules:\n    - rule_id: R1\n      statement: \"Enterprise deals >$250K require CFO approval, adding 30+ days to cycle\"\n      derived_from: [P2, P3]\n      \n      structure:\n        condition: \"deal_size > 250000\"\n        prediction: \"sales_cycle += 30 days\"\n        mechanism: \"CFO approval requirement\"\n        \n      applicability:\n        domain: \"Enterprise sales\"\n        segments: [\"all enterprise\"]\n        exceptions: [\"existing customers with MSA\"]\n        \n    - rule_id: R2\n      statement: \"Legal review stalls >21 days predict deal loss with 80% probability\"\n      derived_from: [P1, P4]\n      \n      structure:\n        condition: \"stall_duration > 21 AND stall_stage = 'legal'\"\n        prediction: \"outcome = 'lost' (p=0.80)\"\n        mechanism: \"Budget cycle expiration, champion fatigue\"\n        \n      applicability:\n        domain: \"Enterprise sales\"\n        segments: [\"new customers\"]\n        exceptions: [\"government deals with known long cycles\"]\n        \n    - rule_id: R3\n      statement: \"58% of enterprise deals will stall at legal review\"\n      derived_from: [P1]\n      \n      structure:\n        condition: \"enterprise deal\"\n        prediction: \"P(legal_stall) = 0.58\"\n        mechanism: \"Custom contract requirements\"\n        \n      applicability:\n        domain: \"Enterprise sales\"\n        segments: [\"all\"]\n        exceptions: [\"standard contract accepted\"]\n```\n\n### Stage 4: Confidence Bounds\n\n**Purpose:** Quantify reliability and identify exceptions.\n\n**Confidence Calculation:**\n```\nConfidence = f(sample_size, pattern_strength, consistency, recency)\n\nBase confidence from sample size:\n  N \u003C 5:   max 0.40\n  N 5-10:  max 0.60\n  N 11-20: max 0.80\n  N > 20:  max 0.95\n\nAdjustments:\n  × pattern_strength (0-1)\n  × consistency (0-1)\n  × recency_weight (0.5-1.0)\n```\n\n**Components:**\n```yaml\nconfidence_analysis:\n  rules:\n    - rule_id: R1\n      confidence: 0.72\n      calculation:\n        base: 0.80        # N=12, moderate sample\n        strength: 0.85    # Strong pattern\n        consistency: 0.88 # Good data quality\n        recency: 0.95     # Recent data\n        final: 0.72       # base × min(strength, consistency, recency)\n        \n      bounds:\n        lower: 0.58       # Pessimistic estimate\n        upper: 0.82       # Optimistic estimate\n        \n      exceptions:\n        identified:\n          - \"Existing customer deal closed in 45 days despite $400K size\"\n            explanation: \"Pre-existing MSA eliminated legal review\"\n          - \"Government deal took 180 days but won\"\n            explanation: \"Known government procurement cycle\"\n        exception_rate: 0.17  # 2\u002F12 instances\n        \n      validity:\n        expires: \"2025-06-01\"  # Re-validate after 6 months\n        invalidated_by: \n          - \"Process change eliminating legal review\"\n          - \"New contract template adoption\"\n        strengthened_by:\n          - \"3+ more instances following pattern\"\n          - \"Causal mechanism confirmed\"\n          \n    - rule_id: R2\n      confidence: 0.68\n      # ... similar structure\n```\n\n**Output Summary:**\n```yaml\ninductive_output:\n  summary:\n    rules_generated: 3\n    highest_confidence: R1 (0.72)\n    total_observations: 12\n    time_range: \"Q3-Q4 2024\"\n    \n  actionable_rules:\n    - rule: R1\n      action: \"Add 30 days to forecast for deals >$250K\"\n      confidence: 0.72\n      \n    - rule: R2\n      action: \"Escalate intervention when legal stall exceeds 14 days\"\n      confidence: 0.68\n      \n  tentative_rules:\n    - rule: R3\n      action: \"Plan for legal stall in 60% of deals (resource accordingly)\"\n      confidence: 0.55\n      needs: \"5+ more observations to reach actionable confidence\"\n      \n  canvas_implications:\n    validate:\n      - assumption: \"A4: Enterprise sales cycle is 90 days\"\n        finding: \"Actually 120 days for deals >$250K\"\n        action: \"Update assumption\"\n        \n    new_hypothesis:\n      - \"H17: Standard contract template would reduce legal stalls by 50%\"\n        basis: \"Legal stall is primary cycle driver\"\n        test: \"Pilot standard contract with 5 deals\"\n```\n\n## Quality Gates\n\n| Gate | Requirement | Failure Action |\n|------|-------------|----------------|\n| Sample size | ≥5 instances | Collect more data |\n| Data quality | ≥80% completeness | Clean dataset |\n| Pattern strength | ≥0.6 for at least one | Lower threshold or collect more |\n| Exception rate | \u003C30% for actionable rules | Narrow rule scope |\n| Mechanism identified | Plausible explanation | Add abductive analysis |\n\n## Common Failure Modes\n\n| Failure | Symptom | Fix |\n|---------|---------|-----|\n| **Small N** | High variance, unstable patterns | Wait for more data |\n| **Survivorship bias** | Only successful cases analyzed | Include failures |\n| **Confounding** | Correlation ≠ causation | Test mechanism with intervention |\n| **Overfitting** | Rule too specific to sample | Simplify rule, test holdout |\n| **Recency bias** | Old patterns weighted equally | Apply recency weighting |\n| **Cherry-picking** | Only confirming instances | Systematic collection |\n\n## Pattern Validation\n\nBefore promoting rule to actionable:\n\n### Statistical Validation\n```yaml\nvalidation:\n  method: \"holdout\"\n  training_set: 8 instances\n  test_set: 4 instances\n  rule_accuracy_on_test: 0.75\n  passed: true\n```\n\n### Causal Validation\n```yaml\nvalidation:\n  method: \"mechanism_test\"\n  proposed_mechanism: \"CFO approval adds 30 days\"\n  test: \"Interview 3 CFOs about approval process\"\n  result: \"Confirmed - CFO review averages 25-35 days\"\n  passed: true\n```\n\n### Temporal Validation\n```yaml\nvalidation:\n  method: \"stability_check\"\n  pattern_in_Q3: 0.62\n  pattern_in_Q4: 0.54\n  drift: -0.08 (acceptable)\n  passed: true\n```\n\n## Automated Pattern Detection\n\nFor continuous learning, run inductive scans:\n\n```yaml\nautomated_scan:\n  frequency: weekly\n  sources:\n    - \"threads\u002Fsales\u002F*\u002F6-learning.md\"\n    - \"threads\u002Fmarketing\u002F*\u002F6-learning.md\"\n    - \"threads\u002Foperations\u002F*\u002F6-learning.md\"\n    \n  thresholds:\n    min_instances: 5\n    min_pattern_strength: 0.6\n    \n  output:\n    location: \"ops\u002Fpatterns.md\"\n    alert_threshold: 0.75  # Flag high-confidence new patterns\n    \n  actions:\n    new_pattern_detected: \"Flag in ops\u002Ftoday.md for review\"\n    existing_pattern_strengthened: \"Update confidence, log\"\n    pattern_invalidated: \"Alert, review rule\"\n```\n\n## Output Contract\n\n```yaml\ninductive_output:\n  observations:\n    count: int\n    time_range: string\n    sources: [string]\n    quality_score: float\n    \n  patterns:\n    detected: [{\n      pattern_id: string\n      type: frequency | correlation | sequence | cluster | trend | threshold\n      description: string\n      strength: float\n      evidence: string\n    }]\n    rejected: [{pattern: string, reason: string}]\n    \n  rules:\n    - rule_id: string\n      statement: string\n      confidence: float\n      bounds: {lower: float, upper: float}\n      exceptions: [{instance: string, explanation: string}]\n      applicability: {domain: string, segments: [string], exceptions: [string]}\n      validity: {expires: date, invalidated_by: [string]}\n      \n  canvas_implications:\n    validate: [{assumption: string, finding: string, action: string}]\n    invalidate: [{assumption: string, finding: string, action: string}]\n    new_hypotheses: [{hypothesis: string, basis: string, test: string}]\n    \n  actions:\n    immediate: [string]        # High-confidence rules to act on\n    monitor: [string]          # Tentative patterns to watch\n    collect: [string]          # Data gaps to fill\n    \n  next:\n    suggested_mode: ReasoningMode\n    threads_to_create: [string]\n    \n  trace:\n    patterns_evaluated: int\n    rules_generated: int\n    duration_ms: int\n```\n\n## Example Execution\n\n**Context:** \"Review last 6 months of marketing content performance\"\n\n**Stage 1 - Observations:**\n```\nCollected: 24 content pieces\nSources: threads\u002Fmarketing\u002F*\u002F6-learning.md\nAttributes: topic, format, channel, sessions, conversions, time_to_demo\nQuality: 0.88 completeness\n```\n\n**Stage 2 - Patterns:**\n```\nP1 (frequency): Case studies convert 2.3x average (8\u002F24, all above average)\nP2 (correlation): Technical depth correlates with enterprise demos (r=0.68)\nP3 (threshold): Posts >2000 words perform better on SEO (breakpoint identified)\nP4 (trend): LinkedIn declining, organic search rising over 6 months\n```\n\n**Stage 3 - Generalizations:**\n```\nR1: \"Case studies should be prioritized for bottom-funnel conversion\"\n    Confidence: 0.75, based on 8 instances\n\nR2: \"Technical content attracts enterprise prospects\"\n    Confidence: 0.68, based on correlation analysis\n\nR3: \"SEO content should target >2000 words\"\n    Confidence: 0.70, based on threshold analysis\n```\n\n**Stage 4 - Confidence Bounds:**\n```\nR1: 0.75 [0.62, 0.85] - Actionable\nR2: 0.68 [0.54, 0.78] - Actionable with caution\nR3: 0.70 [0.58, 0.80] - Actionable\n\nCanvas update: \n  - Validate H8 (case studies convert)\n  - New H18: \"Long-form SEO content drives organic growth\"\n  \nAction: Shift content mix toward case studies and long-form technical guides\n```\n","# Inductive Reasoning\n\nGeneralize from instances to rules. The logic of pattern extraction and empirical learning.\n\n## Type Signature\n\n```\nInductive : [Observation] → Pattern → Generalization → ConfidenceBounds\n\nWhere:\n  Observations     : [Instance] → Dataset\n  Pattern          : Dataset → (Regularity × Frequency)\n  Generalization   : (Regularity × Frequency) → Rule\n  ConfidenceBounds : Rule × SampleSize → (Confidence × Exceptions)\n```\n\n## When to Use\n\n**Use inductive when:**\n- Multiple similar observations accumulate\n- Looking for recurring patterns across threads\n- Building predictive rules from experience\n- Identifying systemic behaviors\n- Validating or discovering Canvas assumptions\n- \"This keeps happening\" situations\n\n**Don't use when:**\n- Explaining single observation → Use Abductive\n- Known causal chain exists → Use Causal\n- Transferring one case to another → Use Analogical\n- Resolving disagreement → Use Dialectical\n\n## Distinction from Other Modes\n\n| Mode | Input | Output | Question |\n|------|-------|--------|----------|\n| **Abductive** | Single anomaly | Explanation | \"Why did this happen?\" |\n| **Inductive** | Multiple instances | Pattern\u002FRule | \"What keeps happening?\" |\n| **Analogical** | One source case | Transferred solution | \"How is this like that?\" |\n\n**Key difference from Abductive:**\n- Abductive: 1 observation → 1 explanation\n- Inductive: N observations → 1 generalization\n\n## Four-Stage Process\n\n### Stage 1: Observation Collection\n\n**Purpose:** Gather and structure multiple instances for analysis.\n\n**Minimum Sample Requirements:**\n\n| Confidence Target | Minimum N | Notes |\n|-------------------|-----------|-------|\n| Exploratory | 3-5 | Hypothesis generation only |\n| Tentative | 6-10 | Directional confidence |\n| Moderate | 11-20 | Actionable patterns |\n| High | 21+ | Strong generalizations |\n\n**Components:**\n```yaml\nobservations:\n  dataset:\n    - instance_id: \"deal-001\"\n      timestamp: ISO8601\n      context: \"Enterprise sales\"\n      attributes:\n        deal_size: 400000\n        sales_cycle: 120\n        stalled_at: \"legal_review\"\n        outcome: \"won\"\n        \n    - instance_id: \"deal-002\"\n      timestamp: ISO8601\n      context: \"Enterprise sales\"\n      attributes:\n        deal_size: 350000\n        sales_cycle: 150\n        stalled_at: \"legal_review\"\n        outcome: \"lost\"\n        \n    # ... more instances\n    \n  metadata:\n    total_instances: 12\n    time_range: \"Q3-Q4 2024\"\n    source: \"threads\u002Fsales\u002F*\u002F6-learning.md\"\n    collection_method: \"automated scan\"\n    \n  quality:\n    completeness: 0.92  # % of fields populated\n    consistency: 0.88   # % following same schema\n    recency: 0.75       # Weight toward recent\n```\n\n### Stage 2: Pattern Detection\n\n**Purpose:** Identify regularities in the dataset.\n\n**Pattern Types:**\n\n| Type | Description | Example |\n|------|-------------|---------|\n| **Frequency** | How often X occurs | \"7\u002F12 deals stall at legal\" |\n| **Correlation** | X and Y co-occur | \"Large deals AND long cycles\" |\n| **Sequence** | X follows Y | \"Stall → lose within 30 days\" |\n| **Cluster** | Groups emerge | \"Two deal archetypes exist\" |\n| **Trend** | Direction over time | \"Cycles getting longer\" |\n| **Threshold** | Breakpoint exists | \"Deals >$300K behave differently\" |\n\n**Detection Process:**\n```yaml\npatterns:\n  detected:\n    - pattern_id: P1\n      type: frequency\n      description: \"Legal review stalls\"\n      evidence: \"7 of 12 deals (58%) stalled at legal review\"\n      strength: 0.78\n      \n    - pattern_id: P2\n      type: correlation\n      description: \"Deal size correlates with cycle length\"\n      evidence: \"r=0.72 between deal_size and sales_cycle\"\n      strength: 0.72\n      \n    - pattern_id: P3\n      type: threshold\n      description: \"CFO involvement threshold\"\n      evidence: \"Deals >$250K require CFO, adding 30+ days\"\n      strength: 0.85\n      \n    - pattern_id: P4\n      type: sequence\n      description: \"Stall duration predicts outcome\"\n      evidence: \"Stalls >21 days → 80% loss rate\"\n      strength: 0.80\n      \n  rejected:\n    - pattern: \"Industry affects outcome\"\n      reason: \"No significant difference across industries (p>0.3)\"\n      \n  insufficient_data:\n    - pattern: \"Seasonality effects\"\n      reason: \"Only 2 quarters of data, need 4+ for seasonality\"\n```\n\n### Stage 3: Generalization\n\n**Purpose:** Form rules from validated patterns.\n\n**Rule Formation:**\n```yaml\ngeneralizations:\n  rules:\n    - rule_id: R1\n      statement: \"Enterprise deals >$250K require CFO approval, adding 30+ days to cycle\"\n      derived_from: [P2, P3]\n      \n      structure:\n        condition: \"deal_size > 250000\"\n        prediction: \"sales_cycle += 30 days\"\n        mechanism: \"CFO approval requirement\"\n        \n      applicability:\n        domain: \"Enterprise sales\"\n        segments: [\"all enterprise\"]\n        exceptions: [\"existing customers with MSA\"]\n        \n    - rule_id: R2\n      statement: \"Legal review stalls >21 days predict deal loss with 80% probability\"\n      derived_from: [P1, P4]\n      \n      structure:\n        condition: \"stall_duration > 21 AND stall_stage = 'legal'\"\n        prediction: \"outcome = 'lost' (p=0.80)\"\n        mechanism: \"Budget cycle expiration, champion fatigue\"\n        \n      applicability:\n        domain: \"Enterprise sales\"\n        segments: [\"new customers\"]\n        exceptions: [\"government deals with known long cycles\"]\n        \n    - rule_id: R3\n      statement: \"58% of enterprise deals will stall at legal review\"\n      derived_from: [P1]\n      \n      structure:\n        condition: \"enterprise deal\"\n        prediction: \"P(legal_stall) = 0.58\"\n        mechanism: \"Custom contract requirements\"\n        \n      applicability:\n        domain: \"Enterprise sales\"\n        segments: [\"all\"]\n        exceptions: [\"standard contract accepted\"]\n```\n\n### Stage 4: Confidence Bounds\n\n**Purpose:** Quantify reliability and identify exceptions.\n\n**Confidence Calculation:**\n```\nConfidence = f(sample_size, pattern_strength, consistency, recency)\n\nBase confidence from sample size:\n  N \u003C 5:   max 0.40\n  N 5-10:  max 0.60\n  N 11-20: max 0.80\n  N > 20:  max 0.95\n\nAdjustments:\n  × pattern_strength (0-1)\n  × consistency (0-1)\n  × recency_weight (0.5-1.0)\n```\n\n**Components:**\n```yaml\nconfidence_analysis:\n  rules:\n    - rule_id: R1\n      confidence: 0.72\n      calculation:\n        base: 0.80        # N=12, moderate sample\n        strength: 0.85    # Strong pattern\n        consistency: 0.88 # Good data quality\n        recency: 0.95     # Recent data\n        final: 0.72       # base × min(strength, consistency, recency)\n        \n      bounds:\n        lower: 0.58       # Pessimistic estimate\n        upper: 0.82       # Optimistic estimate\n        \n      exceptions:\n        identified:\n          - \"Existing customer deal closed in 45 days despite $400K size\"\n            explanation: \"Pre-existing MSA eliminated legal review\"\n          - \"Government deal took 180 days but won\"\n            explanation: \"Known government procurement cycle\"\n        exception_rate: 0.17  # 2\u002F12 instances\n        \n      validity:\n        expires: \"2025-06-01\"  # Re-validate after 6 months\n        invalidated_by: \n          - \"Process change eliminating legal review\"\n          - \"New contract template adoption\"\n        strengthened_by:\n          - \"3+ more instances following pattern\"\n          - \"Causal mechanism confirmed\"\n          \n    - rule_id: R2\n      confidence: 0.68\n      # ... similar structure\n```\n\n**Output Summary:**\n```yaml\ninductive_output:\n  summary:\n    rules_generated: 3\n    highest_confidence: R1 (0.72)\n    total_observations: 12\n    time_range: \"Q3-Q4 2024\"\n    \n  actionable_rules:\n    - rule: R1\n      action: \"Add 30 days to forecast for deals >$250K\"\n      confidence: 0.72\n      \n    - rule: R2\n      action: \"Escalate intervention when legal stall exceeds 14 days\"\n      confidence: 0.68\n      \n  tentative_rules:\n    - rule: R3\n      action: \"Plan for legal stall in 60% of deals (resource accordingly)\"\n      confidence: 0.55\n      needs: \"5+ more observations to reach actionable confidence\"\n      \n  canvas_implications:\n    validate:\n      - assumption: \"A4: Enterprise sales cycle is 90 days\"\n        finding: \"Actually 120 days for deals >$250K\"\n        action: \"Update assumption\"\n        \n    new_hypothesis:\n      - \"H17: Standard contract template would reduce legal stalls by 50%\"\n        basis: \"Legal stall is primary cycle driver\"\n        test: \"Pilot standard contract with 5 deals\"\n```\n\n## Quality Gates\n\n| Gate | Requirement | Failure Action |\n|------|-------------|----------------|\n| Sample size | ≥5 instances | Collect more data |\n| Data quality | ≥80% completeness | Clean dataset |\n| Pattern strength | ≥0.6 for at least one | Lower threshold or collect more |\n| Exception rate | \u003C30% for actionable rules | Narrow rule scope |\n| Mechanism identified | Plausible explanation | Add abductive analysis |\n\n## Common Failure Modes\n\n| Failure | Symptom | Fix |\n|---------|---------|-----|\n| **Small N** | High variance, unstable patterns | Wait for more data |\n| **Survivorship bias** | Only successful cases analyzed | Include failures |\n| **Confounding** | Correlation ≠ causation | Test mechanism with intervention |\n| **Overfitting** | Rule too specific to sample | Simplify rule, test holdout |\n| **Recency bias** | Old patterns weighted equally | Apply recency weighting |\n| **Cherry-picking** | Only confirming instances | Systematic collection |\n\n## Pattern Validation\n\nBefore promoting rule to actionable:\n\n### Statistical Validation\n```yaml\nvalidation:\n  method: \"holdout\"\n  training_set: 8 instances\n  test_set: 4 instances\n  rule_accuracy_on_test: 0.75\n  passed: true\n```\n\n### Causal Validation\n```yaml\nvalidation:\n  method: \"mechanism_test\"\n  proposed_mechanism: \"CFO approval adds 30 days\"\n  test: \"Interview 3 CFOs about approval process\"\n  result: \"Confirmed - CFO review averages 25-35 days\"\n  passed: true\n```\n\n### Temporal Validation\n```yaml\nvalidation:\n  method: \"stability_check\"\n  pattern_in_Q3: 0.62\n  pattern_in_Q4: 0.54\n  drift: -0.08 (acceptable)\n  passed: true\n```\n\n## Automated Pattern Detection\n\nFor continuous learning, run inductive scans:\n\n```yaml\nautomated_scan:\n  frequency: weekly\n  sources:\n    - \"threads\u002Fsales\u002F*\u002F6-learning.md\"\n    - \"threads\u002Fmarketing\u002F*\u002F6-learning.md\"\n    - \"threads\u002Foperations\u002F*\u002F6-learning.md\"\n    \n  thresholds:\n    min_instances: 5\n    min_pattern_strength: 0.6\n    \n  output:\n    location: \"ops\u002Fpatterns.md\"\n    alert_threshold: 0.75  # Flag high-confidence new patterns\n    \n  actions:\n    new_pattern_detected: \"Flag in ops\u002Ftoday.md for review\"\n    existing_pattern_strengthened: \"Update confidence, log\"\n    pattern_invalidated: \"Alert, review rule\"\n```\n\n## Output Contract\n\n```yaml\ninductive_output:\n  observations:\n    count: int\n    time_range: string\n    sources: [string]\n    quality_score: float\n    \n  patterns:\n    detected: [{\n      pattern_id: string\n      type: frequency | correlation | sequence | cluster | trend | threshold\n      description: string\n      strength: float\n      evidence: string\n    }]\n    rejected: [{pattern: string, reason: string}]\n    \n  rules:\n    - rule_id: string\n      statement: string\n      confidence: float\n      bounds: {lower: float, upper: float}\n      exceptions: [{instance: string, explanation: string}]\n      applicability: {domain: string, segments: [string], exceptions: [string]}\n      validity: {expires: date, invalidated_by: [string]}\n      \n  canvas_implications:\n    validate: [{assumption: string, finding: string, action: string}]\n    invalidate: [{assumption: string, finding: string, action: string}]\n    new_hypotheses: [{hypothesis: string, basis: string, test: string}]\n    \n  actions:\n    immediate: [string]        # High-confidence rules to act on\n    monitor: [string]          # Tentative patterns to watch\n    collect: [string]          # Data gaps to fill\n    \n  next:\n    suggested_mode: ReasoningMode\n    threads_to_create: [string]\n    \n  trace:\n    patterns_evaluated: int\n    rules_generated: int\n    duration_ms: int\n```\n\n## Example Execution\n\n**Context:** \"Review last 6 months of marketing content performance\"\n\n**Stage 1 - Observations:**\n```\nCollected: 24 content pieces\nSources: threads\u002Fmarketing\u002F*\u002F6-learning.md\nAttributes: topic, format, channel, sessions, conversions, time_to_demo\nQuality: 0.88 completeness\n```\n\n**Stage 2 - Patterns:**\n```\nP1 (frequency): Case studies convert 2.3x average (8\u002F24, all above average)\nP2 (correlation): Technical depth correlates with enterprise demos (r=0.68)\nP3 (threshold): Posts >2000 words perform better on SEO (breakpoint identified)\nP4 (trend): LinkedIn declining, organic search rising over 6 months\n```\n\n**Stage 3 - Generalizations:**\n```\nR1: \"Case studies should be prioritized for bottom-funnel conversion\"\n    Confidence: 0.75, based on 8 instances\n\nR2: \"Technical content attracts enterprise prospects\"\n    Confidence: 0.68, based on correlation analysis\n\nR3: \"SEO content should target >2000 words\"\n    Confidence: 0.70, based on threshold analysis\n```\n\n**Stage 4 - Confidence Bounds:**\n```\nR1: 0.75 [0.62, 0.85] - Actionable\nR2: 0.68 [0.54, 0.78] - Actionable with caution\nR3: 0.70 [0.58, 0.80] - Actionable\n\nCanvas update: \n  - Validate H8 (case studies convert)\n  - New H18: \"Long-form SEO content drives organic growth\"\n  \nAction: Shift content mix toward case studies and long-form technical guides\n```\n",{"name":5,"description":7},"main","def8531e65114c0fca8fb8551c1871ee0eed705c","skills\u002Faiskillstore\u002Fmarketplace\u002Fbellabe\u002Freasoning-inductive",24,"https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub","https:\u002F\u002Fskillshub.wtf","https:\u002F\u002Fgithub.com\u002FComeOnOliver","AI & Blockchain Engineer\r\n","2026-06-24 22:07:19",{"source":8,"skillName":5,"commands":43,"prompt":53,"zipAvailable":54},[44,47,50],{"tool":45,"command":46},"npx","npx skills add ComeOnOliver\u002Fskillshub --skill reasoning-inductive",{"tool":48,"command":49},"bunx","bunx skills add ComeOnOliver\u002Fskillshub --skill reasoning-inductive",{"tool":51,"command":52},"pnpm","pnpm dlx skills add ComeOnOliver\u002Fskillshub --skill reasoning-inductive","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Faiskillstore\u002Fmarketplace\u002Fbellabe\u002Freasoning-inductive\u002FSKILL.md",true,[56],{"path":57,"resourceType":58,"resourceTypeDesc":59,"sizeBytes":60,"sourceHtmlUrl":61},"skill-report.json","REFERENCE","参考文档",9686,"https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Faiskillstore\u002Fmarketplace\u002Fbellabe\u002Freasoning-inductive\u002Fskill-report.json",[63,77,87,96,102,109],{"skillNo":64,"name":65,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":66,"categories":67,"keywords":73,"publishedAt":28},"01M6FKJN13GMG4KY57J3QD8GY5","model-hierarchy-skill","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzscole\u002Fmodel-hierarchy-skill\u002Fmodel-hierarchy-skill\u002FSKILL.md",[68],{"categoryNo":69,"categoryName":70,"taxonomyType":71,"taxonomyTypeDesc":72},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[74,75,76],"model","model hierarchy","model hierarchy skill",{"skillNo":78,"name":79,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":83,"publishedAt":28},"01M6FKJJJ6GCAMR4PPPRD12M5X","zoom-mcp","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzoom\u002Fskills\u002Fzoom-mcp\u002FSKILL.md",[82],{"categoryNo":69,"categoryName":70,"taxonomyType":71,"taxonomyTypeDesc":72},[84,85,86],"MCP","zoom","zoom mcp",{"skillNo":88,"name":89,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":28},"01M6FKJE3KYT3KPEZMEEHYKA6K","zoom-apps-sdk","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzoom\u002Fskills\u002Fzoom-apps-sdk\u002FSKILL.md",[],[93,94,95,85],"sdk","zoom apps sdk","apps",{"skillNo":97,"name":98,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":99,"categories":100,"keywords":101,"publishedAt":28},"01M6FKJ9EDH9Q5VWDF8HAJRSP0","windows","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzoom\u002Fskills\u002Fwindows\u002FSKILL.md",[],[98],{"skillNo":103,"name":104,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":105,"categories":106,"keywords":108,"publishedAt":28},"01M6FKJ666BEEFC3AE9GJ72Q93","whiteboard","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzoom\u002Fskills\u002Fwhiteboard\u002FSKILL.md",[107],{"categoryNo":69,"categoryName":70,"taxonomyType":71,"taxonomyTypeDesc":72},[104],{"skillNo":110,"name":111,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":112,"categories":113,"keywords":114,"publishedAt":28},"01M6FKJ21WFCDGTJQ2Q91T9DVW","websockets","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzoom\u002Fskills\u002Fwebsockets\u002FSKILL.md",[],[111]]