33GOD Development Lifecycle

Overview

This skill provides meta-level orchestration for the 33GOD event-driven platform, coordinating development across multiple independent components using hierarchical BMAD workflows and Bloodbank events.

The Event-Driven Architecture:

┌─────────────────────────────────────────────────────────────────────────────┐
│                    33GOD PLATFORM ARCHITECTURE                               │
├─────────────────────────────────────────────────────────────────────────────┤
│                                                                              │
│   HOLYFIELDS → BLOODBANK → CANDYSTORE → HOLOCENE → COMPONENTS               │
│   (Schemas)     (Events)    (History)    (View)      (Services)             │
│                    │                              & AGENTS                  │
│                    │                                                         │
│                    ▼                                                         │
│            system.heartbeat.tick                                             │
│            (every 60 seconds)                                                │
│                    │                                                         │
│                    ▼                                                         │
│            HeartbeatRouter                                                   │
│                    │                                                         │
│                    ▼                                                         │
│            Agent Coordination                                                │
│            & Cross-Component Sync                                           │
│                                                                              │
│   This skill operates at the PLATFORM level, coordinating all components     │
│   through BMAD workflows AND Bloodbank events.                               │
│                                                                              │
└─────────────────────────────────────────────────────────────────────────────┘

It enables Director of Engineering level coordination, cross-component story management, and component maturity tracking through both traditional workflows and event-driven automation.

Core Concepts

Hierarchical BMAD Structure

The 33GOD platform uses two BMAD levels:

Meta-Level (Platform root at /33GOD):

• Project Level 4 (Enterprise expansion)
• Coordinates cross-component initiatives
• Tracks integration dependencies
• Manages platform roadmap

Component-Level (Each service directory):

• Project Level 0-3 (varies by component)
• Independent BMAD workflows
• Component-specific development
• Reports status to meta-level

See references/hierarchical-bmad.md for detailed architecture.

Delegation Model

Pattern: Hierarchical command delegation via Zellij

1. Meta-orchestrator identifies cross-component work
2. Decomposes into component-specific tasks
3. Delegates via Zellij tabs (parallel execution)
4. Monitors completion through status aggregation
5. Verifies integration end-to-end

See references/delegation-patterns.md for patterns and examples.

Workflow Decision Tree

Is this cross-component work?
├─ YES → Use meta-level orchestration
│   ├─ Check platform status (workflow below)
│   ├─ Create platform story (workflow below)
│   └─ Delegate to components (workflow below)
│
└─ NO → Use component-level BMAD
    └─ Navigate to component directory
    └─ Run component orchestrator

Primary Workflows

1. Platform Status Review

Trigger: User asks "/platform-status" or "show me component status"

Purpose: Aggregate BMAD status across all components

Steps:

1. Run platform status script:

   bashCopy

   cd /33GOD
   PLATFORM_ROOT=/33GOD ./scripts/platform-status.sh
   

2. Script scans all component directories for:

   • bmad/config.yaml (component configuration)
   • docs/bmm-workflow-status.yaml (workflow completion)

3. Generates docs/platform-status.md with:

   • Component-by-component status summary
   • Maturity indicators (✓, ⚠, →, -)
   • Cross-component recommendations
   • Integration readiness assessment

4. Present summary to user with key insights:

   • How many components have completed planning
   • How many are in implementation
   • Critical blockers or gaps

Output: Platform status report with actionable recommendations

2. Component Maturity Assessment

Trigger: User asks "/component-inventory" or "assess component maturity"

Purpose: Generate comprehensive maturity matrix for all components

Steps:

1. Run component inventory script:

   bashCopy

   cd /33GOD
   PLATFORM_ROOT=/33GOD ./scripts/component-inventory.sh
   

2. Script evaluates each component across dimensions:

   • BMAD initialization (1 point)
   • Planning artifacts (2 points)
   • Architecture docs (2 points)
   • Test coverage (2 points)
   • Documentation (1 point)
   • Production readiness (1 point)

3. Assigns maturity levels:

   • 🔴 Early (0-1): No BMAD, experimental
   • 🟠 Emerging (2-4): BMAD init, planning started
   • 🟡 Developing (5-7): Active development, partial tests
   • 🟢 Mature (8-9): Production-ready, complete

4. Generates docs/component-inventory.md with:

   • Maturity matrix table
   • Component-by-component details
   • Recommendations by maturity level

5. Present summary with strategic insights

Output: Component inventory report with maturity roadmap

See references/component-maturity-model.md for scoring criteria.

3. Cross-Component Story Creation

Trigger: User describes a feature spanning multiple components

Purpose: Create platform-level story and delegate to component teams

Steps:

1. Use platform story template:

   bashCopy

   cp assets/platform-story-template.md docs/stories/{story-name}.md
   

2. Populate story template with:

   • User story (who, what, why)
   • Component breakdown (specific tasks per component)
   • Integration requirements (how components interact)
   • Deployment strategy (rollout sequence)
   • Rollback plan (failure recovery)

3. Review story with user for completeness:

   • Are all affected components identified?
   • Are dependencies clear?
   • Is integration testing defined?

4. Delegate to components (next workflow)

Example Story Types:

• Schema evolution (event/API contract changes)
• New infrastructure (add shared service)
• Performance optimization (cross-component improvements)
• Security enhancement (unified auth, tracing)

See references/cross-component-stories.md for templates and examples.

4. Story Delegation to Components

Trigger: Platform story approved and ready for implementation

Purpose: Delegate component-specific tasks via Zellij coordination

Steps:

1. Determine delegation approach:

   • Parallel: Components can work independently
   • Sequential: Dependencies require ordering

2. Run delegation script:

   bashCopy

   cd /33GOD
   ./scripts/delegate-story.sh \
     --story docs/stories/{story-name}.md \
     --components "comp1,comp2,comp3"
   

3. Script actions:

   • Creates task file in each component: {component}/docs/delegated-task-{timestamp}.md
   • Opens/switches to Zellij tabs (one per component)
   • Displays task context in each tab
   • Prompts next steps for component orchestrators

4. Each component tab shows:

   Copy

   📋 Delegated task ready: docs/delegated-task-{timestamp}.md
   
   Run: /workflow-status to check current state
   Run: /create-story to create implementation story
   

5. Monitor progress:

   • Option A: Attach to Zellij session to watch real-time
   • Option B: Periodic status checks via platform-status.sh

6. Track completion in docs/delegated-stories-tracking.md

Zellij Session Management:

# Attach to platform session
zellij attach 33god-platform

# Switch between component tabs to monitor progress
# Tab names = component names (bloodbank, flume, imi, etc)

See references/delegation-patterns.md for detailed patterns.

5. Integration Map Generation

Trigger: User asks "/integration-map" or "show me component interactions"

Purpose: Visualize component architecture and data flows

Steps:

1. Run integration map script:

   bashCopy

   cd /33GOD
   PLATFORM_ROOT=/33GOD ./scripts/integration-map.sh
   

2. Script analyzes:

   • docker-compose.yml (service dependencies)
   • package.json (workspace dependencies)
   • pyproject.toml (Python dependencies)
   • Component types (frontend, backend, event backbone, etc)

3. Generates docs/integration-map.md with:

   • Mermaid architecture diagram
   • Component registry (type, purpose, dependencies)
   • Event flow diagrams (if event backbone detected)
   • Integration patterns and guidelines

4. Present visualization to user

When to Update:

• New component added to platform
• Integration patterns change
• Architecture review requested
• Onboarding new team members

iMi Cluster Detection

Critical: All 33GOD components follow iMi worktree conventions. Scripts detect components by looking for .iMi/ directories (iMi cluster markers), NOT just any subdirectory.

iMi Cluster Structure

/home/delorenj/code/33GOD/
├── bloodbank/           # iMi cluster
│   ├── .iMi/           # Cluster marker (presence = iMi managed)
│   │   ├── presence/
│   │   ├── links/
│   │   └── registry.toml
│   ├── bmad/           # BMAD at cluster level (NOT in worktrees)
│   ├── trunk-main/     # Trunk worktree
│   └── feat-xyz/       # Feature worktrees (NOT separate components)

Key Points:

• .iMi/ directory identifies an iMi cluster (the actual component)
• Worktrees (trunk-main, feat-*, etc.) are NOT separate components
• BMAD lives at cluster level, shared by all worktrees
• Scripts scan for .iMi/ to avoid treating worktrees as components

Multi-Path Configuration

Problem: 33GOD components span multiple directory levels:

• Top-level projects: /home/delorenj/code/ProjectName/
• Nested 33GOD components: /home/delorenj/code/33GOD/bloodbank/

Solution: Multi-path search via PLATFORM_SEARCH_ROOTS

Default Search Roots:

SEARCH_ROOTS=(
    "/home/delorenj/code"
    "/home/delorenj/code/33GOD"
)

Override with Environment Variable:

PLATFORM_SEARCH_ROOTS="/path1:/path2:/path3" ./scripts/platform-status.sh

How It Works:

1. Scripts iterate through all search roots
2. Find directories containing .iMi/ (iMi clusters)
3. Check if cluster has bmad/ initialized
4. Aggregate results across all search roots

Why This Matters

Without iMi detection:

• Scripts would treat every subdirectory as a component
• Worktrees (feat-user-auth, fix-bug-123) would appear as separate components
• Status reports would be polluted with non-components

With iMi detection:

• Only true iMi clusters (components) are detected
• Worktrees are correctly ignored
• Clean, accurate component inventory

Automation Scripts

All scripts support iMi cluster detection and multi-path configuration.

platform-status.sh

Aggregates BMAD status from all iMi cluster components

• Input: Multi-path search roots
• Output: docs/platform-status.md
• Usage: ./scripts/platform-status.sh
• Override: PLATFORM_SEARCH_ROOTS="/custom/path" ./scripts/platform-status.sh

Detection Logic:

• Scans for .iMi/ directories across all search roots
• Checks if cluster has bmad/config.yaml
• Reports components with BMAD workflows

component-inventory.sh

Generates maturity matrix for all iMi clusters

• Input: Multi-path search roots
• Output: docs/component-inventory.md
• Scoring: 9-point maturity scale
• Usage: ./scripts/component-inventory.sh

Detection Logic:

• Finds all iMi clusters (.iMi/ present)
• Evaluates maturity regardless of BMAD status
• Assesses: BMAD init, planning, architecture, tests, docs, production config

delegate-story.sh

Delegates tasks to component orchestrators via Zellij

• Input: Story file, component names (comma-separated)
• Output: Task files per component, Zellij tabs
• Usage: ./scripts/delegate-story.sh --story {file} --components "bloodbank,flume,imi"

Component Resolution:

• Takes component names, searches for matching .iMi/ clusters
• Searches across all search roots
• Reports error if component not found with search paths

integration-map.sh

Generates Mermaid architecture diagram for iMi clusters

• Input: Multi-path search roots
• Output: docs/integration-map.md with diagrams
• Usage: ./scripts/integration-map.sh

Detection Logic:

• Finds all iMi clusters
• Analyzes dependencies (docker-compose, package.json, pyproject.toml)
• Generates visual component interaction map

Templates and Assets

Platform Story Template

Location: assets/platform-story-template.md

Comprehensive template for cross-component user stories including:

• User story format
• Component breakdown
• Integration requirements
• Deployment strategy
• Rollback plan

Copy to docs/stories/ when creating new platform story.

Component Registry Template

Location: assets/component-registry-template.md

Template for documenting component inventory with:

• Maturity summary
• Production component details
• Integration points
• Technology stack
• Ownership information

Can be used as starting point for manual registry or reference for script output.

Integration Map Template

Location: assets/integration-map-template.mmd

Mermaid diagram template showing:

• Component types (frontend, backend, event backbone, storage)
• Data flow connections
• Styling by component role

Reference when manually creating architecture diagrams.

Reference Documentation

hierarchical-bmad.md

Detailed explanation of two-level BMAD architecture:

• Meta-level vs component-level responsibilities
• Story propagation patterns (top-down)
• Status aggregation patterns (bottom-up)
• Phase alignment strategies
• Independence vs coordination guidelines

When to read: Understanding hierarchical BMAD structure, designing platform architecture, planning cross-component initiatives.

delegation-patterns.md

Comprehensive guide to delegation workflows:

• Zellij-based coordination patterns
• File-based communication protocols
• Parallel vs sequential delegation
• Error handling and rollback
• Example end-to-end delegation

When to read: Delegating tasks to components, troubleshooting delegation issues, designing custom delegation workflows.

component-maturity-model.md

Framework for assessing component production readiness:

• Maturity levels (Early, Emerging, Developing, Mature)
• Scoring criteria (9-point scale)
• Assessment workflows
• Improvement strategies by maturity level
• Cross-component maturity considerations

When to read: Assessing component readiness, planning maturity improvements, prioritizing platform investments.

cross-component-stories.md

Guide to writing effective platform-level stories:

• Story template with complete example
• Story patterns (schema evolution, infrastructure, optimization)
• Component breakdown structure
• Integration requirements definition
• Anti-patterns to avoid

When to read: Creating cross-component stories, reviewing story quality, teaching story structure to teams.

Best Practices

When to Use Meta-Level Orchestration

Use meta-level BMAD when:

• Feature spans 2+ components
• Integration contracts need coordination
• Breaking changes require migration strategy
• Platform-wide standards or architecture
• Strategic roadmap planning

Use component-level BMAD when:

• Feature contained in single component
• No breaking changes to APIs/events
• Implementation details local to component
• Standard component development

Component Independence

Encourage independence:

• Each component has its own BMAD workflow
• Components make local decisions autonomously
• Meta-level only coordinates integration points

Coordinate when necessary:

• API/event schema changes
• Breaking changes
• Cross-component features
• Shared infrastructure changes

Status Aggregation Frequency

Regular checks (recommended):

• Run platform-status.sh weekly or per sprint
• Include in platform status meetings
• Track trends over time

On-demand checks:

• Before major releases
• When planning new initiatives
• During architecture reviews
• When assessing technical debt

Delegation Communication

Clear task boundaries:

• Specify exactly what each component must implement
• Include acceptance criteria per component
• Document dependencies explicitly

Provide full context:

• Include platform story in delegation file
• Explain why the feature matters
• Link related component work

Track completion:

• Use docs/delegated-stories-tracking.md
• Update platform status after completion
• Verify integration end-to-end

Troubleshooting

Component Not Found in Status

Symptom: Script reports "No components with BMAD workflows found"

Resolution:

1. Ensure components have bmad/ directory
2. Check for docs/bmm-workflow-status.yaml in each component
3. Run /workflow-init in component directories missing BMAD

Zellij Tab Not Creating

Symptom: Delegation script fails to create tabs

Resolution:

1. Check if Zellij is installed: which zellij
2. Verify session exists: zellij list-sessions
3. Create session manually: zellij --session 33god-platform
4. Re-run delegation script

Integration Test Failing

Symptom: Components work individually but integration fails

Resolution:

1. Check component-to-component dependencies in integration map
2. Verify API contracts match between producer/consumer
3. Review event schemas for compatibility
4. Check sequence of deployment (dependencies first)
5. Add integration test to catch issues early

Status File Inconsistent

Symptom: Component status doesn't match project level

Resolution:

1. Read component bmad/config.yaml for project level
2. Validate workflow status requirements:
   • Level 0-1: Tech Spec required, PRD recommended
   • Level 2+: PRD required, Tech Spec optional
3. Update status file or config to align
4. Re-run platform-status.sh

Integration with Existing Skills

Use with bmad-orchestrator

The bmad-orchestrator skill handles component-level BMAD workflows. This skill (33god-development-lifecycle) coordinates at the platform level.

Pattern:

User request for cross-component feature
  → This skill: Create platform story, delegate tasks
    → bmad-orchestrator (in each component tab): Run /create-story, /dev-story
      → This skill: Aggregate status, verify integration

Use with zellij-driver

The zellij-driver skill provides programmatic Zellij management. This skill uses Zellij for delegation coordination.

Pattern:

Delegation workflow
  → This skill: Identifies components and tasks
  → zellij-driver: Creates tabs, navigates, executes commands
  → This skill: Monitors completion via status files

Use with 33god-system-expert

The 33god-system-expert skill provides deep architectural knowledge. This skill uses that knowledge for platform planning.

Pattern:

Creating platform architecture
  → This skill: Needs component interaction patterns
  → 33god-system-expert: Provides architectural context
  → This skill: Generates integration map with accurate flows

Example Scenarios

Scenario 1: Add Distributed Tracing

User Request: "Add trace IDs across all components"

Workflow:

1. Run /component-inventory to see affected components
2. Create platform story using template
3. Populate story with:
   • Bloodbank: Add trace_id to event schema
   • Flume: Propagate traces in task execution
   • iMi: Log traces in CLI operations
4. Delegate: ./scripts/delegate-story.sh --story distributed-tracing.md --components "bloodbank,flume,imi"
5. Monitor via Zellij or periodic status checks
6. Verify integration with end-to-end test

Scenario 2: Assess Platform Readiness

User Request: "Are we ready for production?"

Workflow:

1. Run /component-inventory for maturity assessment
2. Review maturity scores:
   • Identify components below Developing level (score <5)
   • Check for missing production configurations
3. Run /platform-status for workflow completion
4. Run /integration-map to verify architecture
5. Create readiness report with gaps and recommendations
6. Prioritize maturity improvements for critical components

Scenario 3: Onboard New Component

User Request: "We're adding a new notifications service"

Workflow:

1. Create component directory: /33GOD/notifications
2. Navigate to component: cd /33GOD/notifications
3. Initialize BMAD: /workflow-init
4. Run /integration-map to update architecture
5. Document integration points in platform architecture
6. Add component to registry
7. Run /platform-status to verify tracking

Meta-Orchestrator Persona

This skill is designed for use by a Director of Engineering persona who:

• Owns platform-wide roadmap
• Coordinates across component teams
• Makes architectural decisions
• Manages cross-component dependencies
• Tracks component maturity and technical debt

Decision Authority:

• Component prioritization
• Breaking change approval
• Architecture evolution
• Platform-wide standards

Delegation Model:

• Defines platform-level requirements
• Delegates implementation to component teams
• Verifies integration quality
• Ensures platform coherence