DocsMIYOMI Launch Readiness Checklist

MIYOMI Launch Readiness Checklist

MIYOMI Launch Readiness Checklist

✅ COMPLETED - Core Infrastructure

  • • [x] Dashboard Buttons: Fixed "Trigger Manual Drop", "Review Pending Picks", "Update Results"
  • • [x] Claude SDK Integration: Full implementation with pick generation and analysis
  • • [x] API Routes: Manual drop, pending picks, update results, status monitoring
  • • [x] Automated Scheduler: 3x daily workflow (11:00, 15:00, 21:00 ET)
  • • [x] Eden Integration: Video generation and registry submission hooks
  • • [x] Workflow Architecture: Complete pick-to-deployment pipeline

    • 🔧 IMMEDIATE NEXT STEPS (Critical)

    1. Environment Configuration

  • • [ ] Set ANTHROPIC_API_KEY for Claude SDK
  • • [ ] Configure EDEN_API_KEY and EDEN_BASE_URL for video generation
  • • [ ] Set INTERNAL_API_TOKEN for secure API access
  • • [ ] Test environment with: npx tsx scripts/start-miyomi-scheduler.ts test

    • 2. Market Data Integration

  • • [ ] Kalshi API: Set up account and API key for market data
  • • [ ] Polymarket API: Configure subgraph access for market resolution
  • • [ ] Manifold API: Add API key for market data and resolution checking
  • • [ ] Implement real market data fetching in update-results/route.ts

    • 3. Database Schema

  • • [ ] Create miyomi_picks table for pick storage and tracking
  • • [ ] Create miyomi_performance table for metrics and results
  • • [ ] Add database persistence to all API routes (currently using mock data)

    • 4. Eden Video Generation

  • • [ ] Confirm Eden API endpoints for MIYOMI video generation
  • • [ ] Test video generation with sample pick and script
  • • [ ] Configure video hosting and CDN for generated content
  • • [ ] Implement video generation status polling

    • 5. Curation Workflow

  • • [ ] Build review interface for pending picks (/dashboard/miyomi/review)
  • • [ ] Implement approval/rejection workflow with trainer feedback
  • • [ ] Add confidence thresholds for auto-approval
  • • [ ] Create pick revision system for improvements

    • 🎯 LAUNCH PHASE (7-14 days)

    Week 1: Alpha Testing

  • • [ ] Deploy scheduler to production environment
  • • [ ] Run 3-5 days of manual drops for testing
  • • [ ] Validate all API integrations and error handling
  • • [ ] Test fallback mechanisms and error recovery
  • • [ ] Verify performance metrics calculations

    • Week 2: Beta Launch

  • • [ ] Enable automated scheduler for daily drops
  • • [ ] Monitor first week of automated performance
  • • [ ] Collect trainer feedback and iterate on pick quality
  • • [ ] Optimize confidence thresholds and curation rules
  • • [ ] Document performance benchmarks

    • 📊 SUCCESS METRICS (30-day targets)

    Launch Gate Validation (Must Hit 3/3)

  • • [ ] Demand Validation: ≥$7,500 gross revenue in 7-day pilot
  • • [ ] Retention Metrics: ≥30% of first-week buyers return within 30 days
  • • [ ] Operational Efficiency: ≥45 saleable outputs/month at ≤$500 compute cost

    • Performance Targets

  • • [ ] Pick accuracy: ≥65% win rate
  • • [ ] Daily drop reliability: ≥95% successful automated drops
  • • [ ] Video generation: ≥90% successful video creation
  • • [ ] User engagement: ≥100 active subscribers within 30 days

    • 🚨 RISK MITIGATION

    Technical Risks

  • • [ ] Claude API Rate Limits: Implement request queuing and fallbacks
  • • [ ] Market API Failures: Add multiple data source redundancy
  • • [ ] Video Generation Delays: Build 24-hour buffer for production
  • • [ ] Database Performance: Monitor query performance under load

    • Quality Risks

  • • [ ] Pick Quality: Implement trainer feedback loop and model improvement
  • • [ ] Market Coverage: Ensure diverse sector representation in picks
  • • [ ] Accuracy Tracking: Real-time performance monitoring and alerts
  • • [ ] Content Moderation: Automated screening for banned topics

    • Business Risks

  • • [ ] Revenue Generation: Track subscription conversions and retention
  • • [ ] Market Demand: Monitor engagement metrics and user feedback
  • • [ ] Competition: Differentiate with contrarian expertise and NYC voice
  • • [ ] Scalability: Ensure infrastructure can handle growth

    • 🔗 INTEGRATION DEPENDENCIES

    Eden Academy Dependencies

  • • [ ] Registry API for work submission
  • • [ ] Media generation pipeline for videos
  • • [ ] Token distribution system for rewards
  • • [ ] Analytics dashboard for performance tracking

    • External API Dependencies

  • • [ ] Anthropic Claude API (critical path)
  • • [ ] Kalshi/Polymarket/Manifold APIs (market data)
  • • [ ] Eden video generation service
  • • [ ] CDN for video hosting and delivery

    • 🚀 LAUNCH COMMANDS

    ``bash

    Environment setup

    cp .env.example .env.local

    Configure all required API keys

    Test the system

    npx tsx scripts/start-miyomi-scheduler.ts test

    Start production scheduler

    npx tsx scripts/start-miyomi-scheduler.ts

    Monitor status

    curl http://localhost:3000/api/miyomi/status

    Manual drop (for testing)

    curl -X POST http://localhost:3000/api/miyomi/manual-drop \ -H "Content-Type: application/json" \ -d '{"agent_id":"miyomi","trigger_time":"2025-08-27T12:00:00Z"}'     ``

    📈 GRADUATION CRITERIA

    MIYOMI will be considered "graduated" and fully autonomous when:

  • Technical Maturity: 30 days of 95%+ automated drop success rate
  • Financial Viability: Consistent $5K+ monthly revenue with positive unit economics
  • Quality Standards: 70%+ pick accuracy rate sustained over 100 picks
  • Community Growth: 500+ active subscribers with 40%+ retention rate
  • Operational Autonomy: Trainer intervention required <5% of the time
    • Next Immediate Action: Configure environment variables and run test workflow Timeline to Launch: 7-14 days with focused development effort Critical Path: Environment setup → Market API integration → Database persistence → Production deployment