Skill Chain Cookbook

Skill Chain Cookbook

Production AI agents from recipes, not from scratch. This is the primary way to use Skene: pick a recipe, copy-paste the chain, and deploy. Skill Chains first — these 36 recipes are the dishes; they’re built from a library of 765 skills (the ingredients). No ML expertise required.

36 Ready-to-Use Recipes

Each recipe in this cookbook is a proven skill chain you can deploy immediately. Every recipe includes:

• Use case — What problem it solves
• Skills — Which skills to chain together
• ROI — Expected time/cost savings
• Step-by-step instructions — Copy-paste examples
• Expected outcomes — What success looks like

📋 Quick Navigation

💼 Sales & Revenue

• Recipe 1: Sales Deal Qualification Pipeline
• Recipe 2: Financial Intelligence Dashboard
• Recipe 3: Product-Led Sales Handoff
• Recipe 4: Competitive Intelligence Automation

🤝 Customer Success & Support

• Recipe 5: Customer Churn Prevention Pipeline
• Recipe 6: AI Support Deflection System
• Recipe 7: Customer Education Platform
• Recipe 8: AI Ops Conversation Pipeline
• Recipe 9: Customer Onboarding Automation
• Recipe 10: Support Ticket Triage & Resolution

🚀 Growth & Marketing

• Recipe 11: Growth Optimization Engine
• Recipe 12: Content Marketing Automation
• Recipe 13: Freemium Conversion Optimization
• Recipe 14: Usage-Based Pricing Engine
• Recipe 15: Community-Led Growth Engine
• Recipe 16: Multi-Platform Content Distribution
• Recipe 17: Pricing & Packaging Optimization
• Recipe 18: Product Analytics Intelligence

🔧 Product & Engineering

• Recipe 19: Developer Experience Onboarding
• Recipe 20: Product Experimentation Engine
• Recipe 21: API Lifecycle Management
• Recipe 22: Security Code Review Automation
• Recipe 23: Superpowers Development Workflow

🎨 Brand & Content

• Recipe 24: Brand Consistency Engine

⚙️ Operations & Compliance

• Recipe 25: Employee Onboarding Automation
• Recipe 26: Data Quality Automation
• Recipe 27: Compliance Automation Hub
• Recipe 28: E-commerce Revenue Optimization
• Recipe 29: Partnership Ecosystem Automation
• Recipe 30: People Ops Talent Intelligence
• Recipe 31: Data Ops Experimentation Pipeline
• Recipe 32: Partnership Deal Flow

🔬 Research & Strategy

• Recipe 33: Scientific Research Synthesis Pipeline
• Recipe 34: Skene Growth Strategy Chain

🌐 Community

• Recipe 35: Community Advocacy Pipeline

💰 FinOps

• Recipe 36: FinOps Standalone Dashboard

Recipe index by domain

Recipe 1: Sales Deal Qualification Pipeline

Use Case: Automatically qualify, score, and route leads with recommended next actions Skills: 5 chained skills Time: 30-60 seconds per lead (was 2-3 hours) ROI: $20K-$40K/month for 50-lead sales team

The Chain

lead_qualification → opportunity_scoring → deal_inspection →
next_best_action → content_recommender

How It Works

1. lead_qualification — Applies MEDDIC/BANT framework, determines if lead is qualified
2. opportunity_scoring — Scores qualified leads on fit, urgency, budget
3. deal_inspection — Analyzes deal health, identifies risks
4. next_best_action — Recommends specific actions for rep
5. content_recommender — Suggests relevant case studies, decks

Setup Instructions

Step 1: Install RevOps Skills

npx skills-directory install --target all --domain revops

Step 2: Configure Lead Qualification Entry Point

{
  "skill": "lead_qualification",
  "input": {
    "leadId": "{{trigger.leadId}}",
    "framework": "MEDDIC",
    "sources": ["CRM", "enrichment_data", "website_activity"]
  },
  "exit_routing": {
    "qualified": "opportunity_scoring",
    "nurture": "nurture_campaign",
    "disqualified": "archive_lead"
  }
}

Step 3: Chain to Opportunity Scoring

{
  "skill": "opportunity_scoring",
  "input": {
    "opportunityId": "{{previous.opportunityId}}",
    "qualificationData": "{{previous.output}}",
    "scoringModel": "weighted"
  },
  "exit_routing": {
    "high_score": "deal_inspection",
    "medium_score": "next_best_action",
    "low_score": "nurture_campaign"
  }
}

Step 4: Add Deal Inspection

{
  "skill": "deal_inspection",
  "input": {
    "dealId": "{{previous.dealId}}",
    "depth": "comprehensive"
  },
  "exit_routing": {
    "healthy": "next_best_action",
    "at_risk": "escalation_manager",
    "blocked": "deal_surgery"
  }
}

Step 5: Route to Next Best Action

{
  "skill": "next_best_action",
  "input": {
    "context": "{{chain.all_previous_outputs}}",
    "repProfile": "{{user.profile}}",
    "priority": "close_deal"
  },
  "exit_routing": {
    "action_recommended": "content_recommender",
    "needs_manager": "escalation_manager"
  }
}

Step 6: Recommend Content

{
  "skill": "content_recommender",
  "input": {
    "dealContext": "{{chain.context}}",
    "buyerPersona": "{{lead.persona}}",
    "dealStage": "{{deal.stage}}"
  },
  "output": "send_to_rep"
}

Testing the Chain

# Test with sample lead data
curl -X POST http://localhost:3000/api/chains/sales-qualification \
  -H "Content-Type: application/json" \
  -d '{
    "leadId": "test-lead-123",
    "leadData": {
      "company": "Acme Corp",
      "revenue": "$50M",
      "employees": 250,
      "industry": "SaaS"
    }
  }'

Expected Output

{
  "qualified": true,
  "score": 85,
  "tier": "high_value",
  "risks": [],
  "nextActions": ["Schedule discovery call", "Send ROI calculator", "Introduce solutions engineer"],
  "recommendedContent": [
    "Case Study: Similar company in SaaS",
    "ROI Calculator: Enterprise tier",
    "Demo Video: Platform overview"
  ],
  "timeline": "Move to demo within 5 days"
}

Monitoring & Metrics

Track these KPIs:

• Lead-to-opportunity rate: Target > 25%
• Time saved per lead: Target 90%+ reduction
• Opportunity quality: Track close rates
• Rep satisfaction: Survey reps on lead quality

Expected Outcomes

• Week 1: Chain deployed, processing 10-20 leads
• Week 2: Reps report 50%+ time savings on qualification
• Month 1: 25%+ lead-to-opportunity conversion rate
• Month 3: $25K+ value delivered in time savings

Recipe 2: Financial Intelligence Dashboard

Use Case: Real-time CFO dashboard with automated board reporting Skills: 5 chained skills Time: Real-time updates vs monthly 40-hour manual process ROI: $50K+/month in finance team time

The Chain

arr_waterfall → burn_rate_monitor → magic_number →
investor_metrics → scenario_planner

How It Works

1. arr_waterfall — Tracks ARR movements (new, expansion, churn, contraction)
2. burn_rate_monitor — Monitors cash burn and runway
3. magic_number — Calculates sales efficiency
4. investor_metrics — Compiles key metrics (CAC, LTV, Rule of 40)
5. scenario_planner — Models "what if" scenarios

Setup Instructions

Step 1: Install FinOps Skills

npx skills-directory install --target all --domain finops

Step 2: Configure ARR Waterfall (Real-time)

{
  "skill": "arr_waterfall",
  "trigger": "data_change",
  "input": {
    "period": "current_month",
    "sources": ["billing", "crm", "contracts"],
    "granularity": "daily"
  },
  "output_to": "burn_rate_monitor"
}

Step 3: Monitor Burn Rate

{
  "skill": "burn_rate_monitor",
  "input": {
    "arrData": "{{previous.arr}}",
    "expenses": "{{accounting.expenses}}",
    "runway_threshold": 12
  },
  "alerts": {
    "runway_below_12_months": "immediate",
    "burn_increasing_20_percent": "warning"
  },
  "output_to": "magic_number"
}

Step 4: Calculate Sales Efficiency

{
  "skill": "magic_number",
  "input": {
    "newARR": "{{arr_waterfall.new}}",
    "salesMarketing": "{{expenses.sales_marketing}}",
    "period": "quarter"
  },
  "benchmark": 1.0,
  "output_to": "investor_metrics"
}

Step 5: Compile Investor Metrics

{
  "skill": "investor_metrics",
  "input": {
    "allData": "{{chain.all_previous}}",
    "includeMetrics": [
      "ARR",
      "Net Revenue Retention",
      "CAC",
      "LTV",
      "LTV:CAC Ratio",
      "Magic Number",
      "Burn Multiple",
      "Rule of 40",
      "Gross Margin"
    ]
  },
  "output_to": "scenario_planner"
}

Step 6: Enable Scenario Planning

{
  "skill": "scenario_planner",
  "input": {
    "baselineData": "{{previous.metrics}}",
    "scenarios": [
      {
        "name": "aggressive_growth",
        "assumptions": { "headcount_increase": 30, "arr_growth": 100 }
      },
      {
        "name": "profitable_growth",
        "assumptions": { "headcount_increase": 15, "arr_growth": 50 }
      }
    ]
  }
}

One-Click Board Report

# Generate board report
npx skills-directory run-chain financial-intelligence \
  --output board-report \
  --format pdf

Expected Outcomes

• Day 1: Dashboard showing real-time metrics
• Week 1: CFO using for daily decision-making
• Month 1: First board report auto-generated
• Quarter 1: 90% reduction in manual financial reporting

Recipe 3: Product-Led Sales Handoff

Use Case: Bridge self-serve and sales motions by automatically routing product-qualified leads to sales with full context Skills: 6 chained skills Time: Automated PQL detection vs manual review (2+ hours per lead) ROI: 3x PQL-to-opportunity conversion, $1M-$3M pipeline capture, 40% faster sales cycles

The Chain

pql_scoring → usage_depth_analyzer → expansion_playbook →
handoff_orchestration → cpq_quote_generator → deal_inspection

How It Works

1. pql_scoring — Identifies product-qualified leads based on usage patterns and engagement
2. usage_depth_analyzer — Analyzes depth of product adoption and power user behaviors
3. expansion_playbook — Recommends expansion strategies based on usage patterns
4. handoff_orchestration — Coordinates seamless handoff from product to sales
5. cpq_quote_generator — Auto-generates quotes based on usage and expansion opportunity
6. deal_inspection — Validates deal health and identifies potential blockers

Setup Instructions

Step 1: Install PLG & RevOps Skills

npx skills-directory install --target all --domain plg revops

Step 2: Configure PQL Scoring

{
  "skill": "pql_scoring",
  "trigger": "scheduled",
  "frequency": "daily",
  "input": {
    "accountIds": "{{all_active_accounts}}",
    "signals": [
      "feature_usage",
      "team_size",
      "integration_count",
      "api_usage",
      "storage_consumption"
    ],
    "threshold": 75,
    "scoringModel": "product_led"
  },
  "exit_routing": {
    "pql_qualified": "usage_depth_analyzer",
    "not_qualified": "nurture_sequence"
  }
}

Step 3: Analyze Usage Depth

{
  "skill": "usage_depth_analyzer",
  "input": {
    "accountId": "{{previous.accountId}}",
    "pqlScore": "{{previous.score}}",
    "analysisDepth": "comprehensive",
    "lookbackWindow": "90_days",
    "metrics": ["feature_breadth", "power_user_count", "collaboration_score"]
  },
  "exit_routing": {
    "expansion_ready": "expansion_playbook",
    "early_stage": "product_education"
  }
}

Step 4: Generate Expansion Playbook

{
  "skill": "expansion_playbook",
  "input": {
    "accountId": "{{chain.accountId}}",
    "usageAnalysis": "{{previous.analysis}}",
    "currentPlan": "{{account.plan}}",
    "expansionVectors": ["seats", "features", "usage_tier", "enterprise_upgrade"],
    "targetARR": "calculate"
  },
  "exit_routing": {
    "playbook_ready": "handoff_orchestration",
    "needs_enrichment": "data_enrichment"
  }
}

Step 5: Orchestrate Sales Handoff

{
  "skill": "handoff_orchestration",
  "input": {
    "accountId": "{{chain.accountId}}",
    "pqlData": "{{chain.pql_data}}",
    "expansionPlan": "{{previous.playbook}}",
    "assignmentRules": "territory_based",
    "notifyRep": true,
    "notifyCustomer": false
  },
  "exit_routing": {
    "assigned_to_sales": "cpq_quote_generator",
    "hold_for_timing": "schedule_outreach"
  }
}

Step 6: Generate Quote

{
  "skill": "cpq_quote_generator",
  "input": {
    "accountId": "{{chain.accountId}}",
    "expansionPlan": "{{chain.expansion_plan}}",
    "discountRules": "apply_pql_discount",
    "pricingModel": "usage_based",
    "includeOnboarding": true
  },
  "exit_routing": {
    "quote_generated": "deal_inspection",
    "needs_approval": "manager_review"
  }
}

Step 7: Inspect Deal Health

{
  "skill": "deal_inspection",
  "input": {
    "dealId": "{{previous.dealId}}",
    "depth": "comprehensive",
    "checkpoints": ["budget", "authority", "timeline", "competition"],
    "riskFactors": true
  },
  "output": "send_to_rep_with_insights"
}

Testing the Chain

curl -X POST http://localhost:3000/api/chains/pql-handoff \
  -H "Content-Type: application/json" \
  -d '{
    "accountId": "test-account-789",
    "usageData": {
      "seats": 15,
      "featuresUsed": 12,
      "apiCalls": 500000,
      "teamCollaboration": "high",
      "powerUsers": 5
    },
    "currentPlan": "pro"
  }'

Expected Output

{
  "pqlQualified": true,
  "pqlScore": 89,
  "usageDepth": "high",
  "expansionOpportunity": {
    "type": "enterprise_upgrade",
    "estimatedARR": "$50,000",
    "confidence": "high"
  },
  "assignedTo": "rep_john_smith",
  "quoteGenerated": true,
  "dealHealth": "healthy",
  "recommendedActions": [
    "Schedule discovery call within 48 hours",
    "Present enterprise features demo",
    "Discuss API rate limits and custom integration needs"
  ],
  "timeline": "Close within 30 days"
}

Monitoring & Metrics

Track these KPIs:

• PQL-to-Opportunity Rate: Target > 60% (baseline ~20%)
• PQL-to-Close Rate: Target > 15% (baseline ~5%)
• Sales Cycle Length: Target 30-40 days (baseline 60-90 days)
• Average Deal Size: Target $50K+ for enterprise upgrades
• Rep Satisfaction: Track rep feedback on PQL quality

Expected Outcomes

• Week 1: PQL detection running, 10-20 accounts identified
• Month 1: 3x improvement in PQL-to-opportunity conversion
• Quarter 1: $1M-$3M incremental pipeline from product-led motion
• Quarter 2: 40% reduction in sales cycle for product-led deals

Recipe 4: Competitive Intelligence Automation

Use Case: Automate competitive research, feature tracking, pricing analysis, and battlecard generation Skills: 6 chained skills Time: Real-time competitive intelligence vs 20-40 hours/month manual research ROI: 90% faster competitive analysis, 60% better win rates, $200K+ pipeline impact

The Chain

competitive_intel → competitive_feature_tracker → ai_knowledge_synthesizer →
content_research_writer → battlecard_generator → win_loss_analyzer

How It Works

1. competitive_intel — Monitors competitors' websites, announcements, and market moves
2. competitive_feature_tracker — Tracks competitive feature releases and updates
3. ai_knowledge_synthesizer — Synthesizes competitive intelligence from multiple sources
4. content_research_writer — Researches and documents competitive positioning
5. battlecard_generator — Creates sales battlecards for competitive situations
6. win_loss_analyzer — Analyzes win/loss patterns against specific competitors

Setup Instructions

Step 1: Install RevOps, Marketing & AI Ops Skills

npx skills-directory install --target all --domain revops marketing ai_ops

Step 2: Configure Competitive Monitoring

{
  "skill": "competitive_intel",
  "trigger": "scheduled",
  "frequency": "daily",
  "input": {
    "competitors": ["competitor_a", "competitor_b", "competitor_c"],
    "monitoringSources": [
      "company_websites",
      "blog_posts",
      "press_releases",
      "product_hunt",
      "linkedin",
      "job_postings",
      "tech_news"
    ],
    "alertCriteria": ["product_launch", "pricing_change", "executive_hire", "funding_news"],
    "sentimentTracking": true
  },
  "exit_routing": {
    "updates_found": "competitive_feature_tracker",
    "no_changes": "continue_monitoring"
  }
}

Step 3: Track Feature Releases

{
  "skill": "competitive_feature_tracker",
  "input": {
    "competitors": "{{previous.competitors}}",
    "competitiveUpdates": "{{previous.updates}}",
    "featureCategories": [
      "core_functionality",
      "integrations",
      "enterprise_features",
      "pricing_tiers",
      "security_compliance"
    ],
    "comparisonMatrix": "auto_update",
    "gapAnalysis": true
  },
  "exit_routing": {
    "features_tracked": "ai_knowledge_synthesizer",
    "critical_feature": "alert_product_team"
  }
}

Step 4: Synthesize Intelligence

{
  "skill": "ai_knowledge_synthesizer",
  "input": {
    "competitiveData": "{{chain.all_data}}",
    "synthesisGoals": [
      "market_positioning",
      "feature_gaps",
      "pricing_strategy",
      "target_customers",
      "go_to_market_approach"
    ],
    "includeTrends": true,
    "confidenceScoring": true,
    "citeSources": true
  },
  "exit_routing": {
    "synthesis_complete": "content_research_writer",
    "needs_validation": "research_team_review"
  }
}

Step 5: Research Competitive Positioning

{
  "skill": "content_research_writer",
  "input": {
    "topic": "competitive_landscape",
    "synthesizedIntel": "{{previous.synthesis}}",
    "outputFormat": "structured_analysis",
    "includeStrengthsWeaknesses": true,
    "targetAudience": "sales_team",
    "updateFrequency": "monthly"
  },
  "exit_routing": {
    "research_complete": "battlecard_generator",
    "needs_expert_input": "analyst_review"
  }
}

Step 6: Generate Sales Battlecards

{
  "skill": "battlecard_generator",
  "input": {
    "competitor": "{{trigger.competitor}}",
    "competitiveIntel": "{{chain.all_intel}}",
    "battlecardSections": [
      "overview",
      "strengths_weaknesses",
      "key_differentiators",
      "common_objections",
      "proof_points",
      "customer_stories",
      "pricing_comparison",
      "talk_tracks"
    ],
    "format": "sales_enablement_tool",
    "updateTrigger": "competitive_change"
  },
  "exit_routing": {
    "battlecard_ready": "win_loss_analyzer",
    "needs_review": "sales_enablement_review"
  }
}

Step 7: Analyze Win/Loss Patterns

{
  "skill": "win_loss_analyzer",
  "input": {
    "competitorId": "{{chain.competitor}}",
    "dealData": "{{crm.closed_deals}}",
    "analysisWindow": "90_days",
    "winLossFactors": [
      "price",
      "features",
      "implementation_time",
      "customer_support",
      "integrations",
      "brand"
    ],
    "includeQuotes": true,
    "recommendationEngine": true
  },
  "output": "win_loss_insights_report"
}

Testing the Chain

curl -X POST http://localhost:3000/api/chains/competitive-intel \
  -H "Content-Type: application/json" \
  -d '{
    "competitor": "competitor_a",
    "monitoringPeriod": "last_7_days",
    "analysisType": "comprehensive"
  }'

Expected Output

{
  "competitiveUpdatesFound": 5,
  "criticalUpdates": 2,
  "featuresTracked": 12,
  "newFeatures": 3,
  "pricingChanges": 1,
  "synthesisGenerated": true,
  "battlecardsUpdated": 3,
  "winRate": {
    "vs_competitor_a": "65%",
    "vs_competitor_b": "72%",
    "vs_competitor_c": "58%"
  },
  "keyInsights": [
    "Competitor A launched enterprise SSO - our feature gap",
    "Competitor B raised prices 20% - opportunity for us",
    "Competitor C struggling with customer support (source: G2)"
  ],
  "recommendations": [
    "Prioritize SSO feature development",
    "Target Competitor B's price-sensitive customers",
    "Emphasize our support quality in competitive situations"
  ]
}

Monitoring & Metrics

Track these KPIs:

• Competitive Updates Detected: Track volume and criticality
• Win Rate by Competitor: Target 60%+ vs each major competitor
• Battlecard Utilization: Track sales team usage
• Time to Intelligence: Target < 24 hours for critical updates
• Competitive Losses: Track and categorize reasons for losses

Expected Outcomes

• Week 1: Competitive monitoring active for 3-5 competitors
• Month 1: 90% faster competitive analysis vs manual research
• Quarter 1: 60% improvement in win rates against tracked competitors
• Quarter 2: $200K+ pipeline impact from competitive insights

Recipe 5: Customer Churn Prevention Pipeline

Use Case: Predict churn risk 60-90 days early and trigger intervention playbooks Skills: 4 chained skills Time: Real-time monitoring vs quarterly reviews ROI: $400K+ ARR saved annually

The Chain

health_scoring → churn_prediction → risk_mitigation_playbook →
escalation_manager

How It Works

1. health_scoring — Continuously monitors account health signals
2. churn_prediction — ML-based prediction of churn risk 60-90 days out
3. risk_mitigation_playbook — Executes tailored intervention strategies
4. escalation_manager — Alerts CSM and triggers manager involvement if needed

Setup Instructions

Step 1: Install Customer Success Skills

npx skills-directory install --target all --domain customer_success

Step 2: Configure Health Scoring (Continuous)

{
  "skill": "health_scoring",
  "trigger": "scheduled",
  "frequency": "daily",
  "input": {
    "accountIds": "{{all_active_accounts}}",
    "signals": [
      "product_usage",
      "support_tickets",
      "nps_score",
      "contract_value",
      "engagement_metrics"
    ]
  },
  "exit_routing": {
    "healthy": "log_and_continue",
    "at_risk": "churn_prediction",
    "critical": "immediate_escalation"
  }
}

Step 3: Chain to Churn Prediction

{
  "skill": "churn_prediction",
  "input": {
    "accountId": "{{previous.accountId}}",
    "healthData": "{{previous.healthScore}}",
    "historicalData": true,
    "predictionWindow": "90_days"
  },
  "exit_routing": {
    "high_risk": "risk_mitigation_playbook",
    "medium_risk": "risk_mitigation_playbook",
    "low_risk": "monitor_only"
  }
}

Step 4: Execute Risk Mitigation

{
  "skill": "risk_mitigation_playbook",
  "input": {
    "accountId": "{{chain.accountId}}",
    "riskLevel": "{{previous.riskScore}}",
    "riskFactors": "{{previous.factors}}",
    "playbookType": "retention"
  },
  "actions": [
    "Schedule executive business review",
    "Conduct product usage analysis",
    "Offer training/onboarding refresh",
    "Introduce customer success resources"
  ],
  "exit_routing": {
    "playbook_started": "escalation_manager",
    "needs_custom_plan": "csm_intervention"
  }
}

Step 5: Escalate if Needed

{
  "skill": "escalation_manager",
  "input": {
    "accountId": "{{chain.accountId}}",
    "context": "{{chain.all_data}}",
    "urgency": "{{previous.riskLevel}}"
  },
  "notifications": ["Assigned CSM", "CSM Manager", "VP Customer Success"]
}

Monitoring Dashboard

Create a real-time dashboard tracking:

┌─────────────────────────────────────────────┐
│ Churn Prevention Dashboard                  │
├─────────────────────────────────────────────┤
│ Accounts Monitored: 347                     │
│ At-Risk (High): 12                          │
│ At-Risk (Medium): 28                        │
│ Playbooks Active: 18                        │
│ ARR at Risk: $450K                          │
│ ARR Saved (QTD): $180K                      │
└─────────────────────────────────────────────┘

Expected Outcomes

• Week 1: Health scoring running on all accounts
• Week 2: First at-risk accounts identified
• Month 1: 3-5 accounts saved from churn
• Quarter 1: $100K-$200K ARR saved

Recipe 6: AI Support Deflection System

Use Case: Automate support ticket triage, classification, and resolution with AI-powered response generation and knowledge gap detection Skills: 6 chained skills Time: < 1 minute automated resolution vs 4-8 hour manual response time ROI: 70% ticket deflection, 50% support cost reduction, $150K-$300K annual savings

The Chain

ai_ticket_classifier → ai_intent_classifier → ai_response_suggester →
support_resolution_suggester → support_kb_gap_finder → support_deflector

How It Works

1. ai_ticket_classifier — Automatically classifies tickets by type, priority, and routing
2. ai_intent_classifier — Identifies customer intent and underlying issue
3. ai_response_suggester — Generates contextual AI-powered response drafts
4. support_resolution_suggester — Recommends resolution steps based on historical patterns
5. support_kb_gap_finder — Identifies missing knowledge base articles
6. support_deflector — Deflects tickets to self-serve resources when appropriate

Setup Instructions

Step 1: Install AI Ops & Support Ops Skills

npx skills-directory install --target all --domain ai_ops support_ops

Step 2: Configure AI Ticket Classification

{
  "skill": "ai_ticket_classifier",
  "trigger": "ticket_created",
  "input": {
    "ticketId": "{{trigger.ticketId}}",
    "ticketContent": "{{trigger.body}}",
    "customerContext": "{{customer.history}}",
    "classificationTypes": ["bug", "feature_request", "how_to", "billing", "technical"],
    "priorityLevels": ["urgent", "high", "normal", "low"]
  },
  "exit_routing": {
    "classified": "ai_intent_classifier",
    "needs_human": "escalate_immediately"
  }
}

Step 3: Classify Customer Intent

{
  "skill": "ai_intent_classifier",
  "input": {
    "ticketId": "{{chain.ticketId}}",
    "ticketType": "{{previous.classification}}",
    "customerMessage": "{{trigger.body}}",
    "contextualData": "{{customer.account_data}}",
    "intentCategories": [
      "resolve_issue",
      "get_information",
      "request_feature",
      "report_bug",
      "cancel_service"
    ]
  },
  "exit_routing": {
    "intent_clear": "ai_response_suggester",
    "intent_ambiguous": "clarification_needed"
  }
}

Step 4: Generate AI Response

{
  "skill": "ai_response_suggester",
  "input": {
    "ticketId": "{{chain.ticketId}}",
    "intent": "{{previous.intent}}",
    "classification": "{{chain.classification}}",
    "knowledgeBase": "search",
    "tone": "professional_friendly",
    "includeLinks": true
  },
  "exit_routing": {
    "response_generated": "support_resolution_suggester",
    "kb_gaps_found": "support_kb_gap_finder"
  }
}

Step 5: Suggest Resolution Steps

{
  "skill": "support_resolution_suggester",
  "input": {
    "ticketId": "{{chain.ticketId}}",
    "intent": "{{chain.intent}}",
    "aiResponse": "{{previous.response}}",
    "historicalResolutions": "search_similar",
    "resolutionType": ["self_serve", "agent_assisted", "escalated"]
  },
  "exit_routing": {
    "self_serve_possible": "support_deflector",
    "needs_agent": "assign_to_agent",
    "needs_escalation": "escalate_to_tier2"
  }
}

Step 6: Find Knowledge Base Gaps

{
  "skill": "support_kb_gap_finder",
  "input": {
    "ticketId": "{{chain.ticketId}}",
    "ticketType": "{{chain.classification}}",
    "searchAttempts": "{{chain.kb_searches}}",
    "commonQueries": "analyze",
    "gapThreshold": 3
  },
  "exit_routing": {
    "gap_found": "create_kb_article_task",
    "no_gap": "support_deflector"
  }
}

Step 7: Deflect to Self-Serve

{
  "skill": "support_deflector",
  "input": {
    "ticketId": "{{chain.ticketId}}",
    "aiResponse": "{{chain.ai_response}}",
    "resolutionSteps": "{{previous.steps}}",
    "kbArticles": "{{chain.relevant_articles}}",
    "deflectionConfidence": "{{previous.confidence}}"
  },
  "output": "send_to_customer_or_agent"
}

Testing the Chain

curl -X POST http://localhost:3000/api/chains/support-deflection \
  -H "Content-Type: application/json" \
  -d '{
    "ticketId": "test-ticket-001",
    "customerMessage": "How do I reset my API key?",
    "customerId": "cust_123",
    "accountType": "pro"
  }'

Expected Output

{
  "ticketClassified": true,
  "classification": "how_to",
  "priority": "normal",
  "intent": "get_information",
  "aiResponseGenerated": true,
  "deflectionPossible": true,
  "kbArticles": ["How to Reset Your API Key", "API Security Best Practices"],
  "resolutionSteps": [
    "Navigate to Settings > API Keys",
    "Click 'Regenerate Key'",
    "Update key in your application"
  ],
  "deflected": true,
  "estimatedResolutionTime": "< 1 minute",
  "supportCostSaved": "$25"
}

Monitoring & Metrics

Track these KPIs:

• Deflection Rate: Target > 70% of eligible tickets
• First Response Time: Target < 1 minute for AI responses
• Customer Satisfaction: Target > 4.5/5 for deflected tickets
• Support Cost Savings: Target $150K-$300K annually
• Knowledge Base Coverage: Track gap identification and article creation

Expected Outcomes

• Week 1: AI classification active on 100% of tickets
• Month 1: 70% deflection rate for how-to and informational tickets
• Quarter 1: $50K-$100K support cost savings
• Quarter 2: 90% coverage in knowledge base, minimal gaps

Recipe 7: Customer Education Platform

Use Case: Automate customer education and product adoption with guided onboarding, milestone celebrations, and content curation Skills: 6 chained skills Time: Automated vs manual 1-on-1 training sessions ROI: 45% higher product adoption, 60% faster time-to-value, 70% reduction in support tickets

The Chain

guided_setup_wizard → milestone_celebration → community_content_curator →
email_sequence → feedback_collection → activation_metrics

How It Works

1. guided_setup_wizard — Walks users through initial product setup step-by-step
2. milestone_celebration — Celebrates user achievements to increase engagement
3. community_content_curator — Surfaces relevant community content and best practices
4. email_sequence — Sends educational email drips based on user progress
5. feedback_collection — Gathers user feedback on educational content
6. activation_metrics — Tracks activation progress and identifies stuck users

Setup Instructions

Step 1: Install Customer Success & PLG Skills

npx skills-directory install --target all --domain customer_success plg

Step 2: Configure Guided Setup

{
  "skill": "guided_setup_wizard",
  "trigger": "user_signup",
  "input": {
    "userId": "{{trigger.userId}}",
    "productComplexity": "medium",
    "setupSteps": [
      {
        "step": 1,
        "title": "Connect your data source",
        "required": true,
        "helpContent": "integration-guide.md"
      },
      {
        "step": 2,
        "title": "Invite your team",
        "required": false,
        "helpContent": "collaboration-guide.md"
      },
      {
        "step": 3,
        "title": "Create your first dashboard",
        "required": true,
        "helpContent": "dashboard-tutorial.md"
      }
    ],
    "adaptiveGuidance": true,
    "allowSkip": false
  },
  "exit_routing": {
    "setup_complete": "milestone_celebration",
    "setup_abandoned": "retention_campaign",
    "stuck": "intervention_trigger"
  }
}

Step 3: Celebrate Milestones

{
  "skill": "milestone_celebration",
  "input": {
    "userId": "{{chain.userId}}",
    "milestone": "{{previous.completed_step}}",
    "celebrationTypes": {
      "in_app_notification": true,
      "confetti_animation": true,
      "badge_award": true,
      "email_notification": true
    },
    "nextMilestone": "auto_suggest",
    "shareOption": "social_media"
  },
  "exit_routing": {
    "celebrated": "community_content_curator",
    "user_inactive": "re_engagement"
  }
}

Step 4: Curate Educational Content

{
  "skill": "community_content_curator",
  "input": {
    "userId": "{{chain.userId}}",
    "userProgress": "{{chain.milestones}}",
    "contentTypes": [
      "video_tutorials",
      "how_to_guides",
      "use_cases",
      "community_posts",
      "webinar_recordings"
    ],
    "personalization": {
      "industry": "{{user.industry}}",
      "useCase": "{{user.use_case}}",
      "skillLevel": "{{user.expertise}}"
    },
    "maxItems": 5
  },
  "exit_routing": {
    "content_delivered": "email_sequence",
    "no_relevant_content": "content_creation_request"
  }
}

Step 5: Send Educational Emails

{
  "skill": "email_sequence",
  "input": {
    "userId": "{{chain.userId}}",
    "sequenceType": "educational_drip",
    "emailSchedule": [
      {
        "day": 1,
        "topic": "Getting Started",
        "content": "{{chain.curated_content[0]}}"
      },
      {
        "day": 3,
        "topic": "Best Practices",
        "content": "{{chain.curated_content[1]}}"
      },
      {
        "day": 7,
        "topic": "Advanced Features",
        "content": "{{chain.curated_content[2]}}"
      }
    ],
    "adaptToProgress": true,
    "includeSuccessStories": true
  },
  "exit_routing": {
    "sequence_started": "feedback_collection",
    "user_unsubscribed": "in_app_only"
  }
}

Step 6: Collect Feedback

{
  "skill": "feedback_collection",
  "input": {
    "userId": "{{chain.userId}}",
    "feedbackTriggers": ["milestone_reached", "content_consumed", "email_opened", "feature_used"],
    "feedbackTypes": {
      "nps": true,
      "content_rating": true,
      "feature_request": true,
      "help_needed": true
    },
    "frequency": "milestone_based",
    "format": "in_app_modal"
  },
  "exit_routing": {
    "feedback_received": "activation_metrics",
    "negative_feedback": "support_escalation"
  }
}

Step 7: Track Activation Metrics

{
  "skill": "activation_metrics",
  "input": {
    "userId": "{{chain.userId}}",
    "activationCriteria": [
      "completed_setup",
      "invited_team_member",
      "created_first_artifact",
      "used_core_feature_3_times",
      "returned_3_days_in_week"
    ],
    "timeWindow": "14_days",
    "activationScore": "calculate",
    "identifyBlockers": true
  },
  "output": "activation_report"
}

Testing the Chain

curl -X POST http://localhost:3000/api/chains/customer-education \
  -H "Content-Type: application/json" \
  -d '{
    "userId": "new-user-456",
    "signupDate": "2026-02-01",
    "industry": "SaaS",
    "useCase": "product_analytics"
  }'

Expected Output

{
  "status": "success",
  "educationPlan": {
    "setupProgress": "67%",
    "milestonesCompleted": 2,
    "milestonesRemaining": 1,
    "contentRecommended": 5,
    "emailsScheduled": 3,
    "feedbackCollected": true,
    "activationScore": 72
  },
  "userEngagement": {
    "setupTime": "12 minutes",
    "contentViews": 3,
    "emailOpenRate": "85%",
    "npsScore": 9,
    "blockers": []
  },
  "nextActions": ["Complete final setup step", "Invite team member", "Explore advanced features"]
}

Monitoring & Metrics

Track these KPIs:

• Activation Rate: Target 60%+ within 14 days (baseline ~30%)
• Time to Value: Target < 2 days (baseline ~7 days)
• Content Engagement: Track views, clicks, completion rates
• Support Ticket Reduction: Target 70% fewer "how do I..." tickets
• NPS from Educated Users: Target 40+ points

Expected Outcomes

• Week 1: Guided setup deployed, 100+ users onboarded
• Month 1: 45% higher activation rate vs. control group
• Quarter 1: 60% faster time-to-value, 70% support ticket reduction
• Quarter 2: $100K+ saved in customer success time

Recipe 8: AI Ops Conversation Pipeline

Use Case: Classify, route, suggest responses, and summarize for support or sales conversations Skills: 4 chained skills (AI Ops) Time: Seconds per ticket; deflected volume ROI: 30–50% deflection, faster first response, consistent summaries

The Chain

ai_ticket_classifier → ai_intent_classifier → ai_response_suggester →
ai_summarization_agent

How It Works

1. ticket_classifier — Categories and prioritizes incoming tickets
2. intent_classifier — Detects user intent for routing and context
3. response_suggester — Suggests replies or next actions
4. summarization_agent — Produces conversation or thread summaries

Setup

Install domain: npx skills-directory install --target all --domain ai_ops. Chain the four skills with exit_routing from each step to the next; final step output to your CRM or support queue.

Expected Outcomes

• Week 1: Pipeline running on sample tickets
• Month 1: 30%+ deflection or faster resolution
• Quarter 1: Summaries used for coaching and reporting

Recipe 9: Customer Onboarding Automation

onboarding_health → guided_setup_wizard → milestone_celebration →
time_to_value → activation_metrics

ROI: 50% faster time-to-value, 30% higher activation

Recipe 10: Support Ticket Triage & Resolution

support_ticket_triage → support_resolution_suggester →
support_kb_gap_finder → support_bug_linker

ROI: 40% faster resolution, 60% ticket deflection

Recipe 11: Growth Optimization Engine

Use Case: Continuous A/B testing and conversion optimization Skills: 6 chained skills Time: 10-15 experiments/month vs 2-4 ROI: 15-25% conversion lift

The Chain

signup_flow_cro → page_cro → ab_test_setup →
analytics_tracking → activation_metrics → feature_adoption

How It Works

1. signup_flow_cro — Optimizes signup conversion
2. page_cro — Optimizes landing page conversion
3. ab_test_setup — Configur

…(truncated)