| id | genesis-orchestrator |
| name | genesis-orchestrator |
| type | skill |
| version | 2.0.1 |
| created | 20/03/2026 |
| modified | 20/03/2026 |
| status | active |
| metadata | {"author":"NodeJS-Starter-V1","version":"2.0.1","locale":"en-AU"} |
| description | Autonomous project orchestration protocol for Next.js full-stack builds. Triggers on \"build\", \"implement\", \"create feature\", \"plan\", \"architecture\", or when starting new phases. Enforces phase-locked execution, token economy, and sectional verification gates.
|
| context | fork |
Genesis Hive Mind Orchestrator
Master orchestration protocol for autonomous Next.js full-stack development. Transforms natural language intent into precise, phase-locked execution commands.
Description
Governs the full lifecycle of feature and project builds by enforcing phase-locked execution, token economy constraints, and sectional verification gates. Decomposes complex requests into Discovery, Vision Board, Blueprint, and Execution phases, delegating to sub-agents (MATH_COUNCIL, TITAN_DESIGN, GENESIS_DEV) as required. Ensures no phase advances without passing all quality checks.
When to Apply
Positive Triggers
- Starting a new feature implementation
- Planning project architecture
- Executing multi-phase development tasks
- Needing to break complex work into verifiable sections
- User says: "build", "implement", "create", "plan", "architecture"
Negative Triggers
- Reviewing or optimising existing code (use
council-of-logic instead)
- Styling or designing UI components (use
scientific-luxury instead)
- Running single-file bug fixes that do not require phased execution
Core Directives
Token Economy Protocol
| Rule | Mechanism | Instruction |
|---|
| NEVER output full project code in one pass | SECTIONAL_EXECUTION | Break every major task into isolated 'Phases'. Complete one phase, verify it, clear context, then proceed. |
Why: Prevents context-window overflow and ensures quality verification at each step.
Sub-Agent Activation
| User Intent | Activate Agent |
|---|
| "optimise", "algorithm", "performance", "logic" | MATH_COUNCIL |
| "design", "UI", "animation", "style", "look" | TITAN_DESIGN |
| "build", "implement", "create", "add feature" | GENESIS_DEV |
| "plan", "architecture", "structure" | BLUEPRINT_MODE |
| "what is", "explain", "how does" | DISCOVERY_MODE |
Autonomous Workflow Loop
PHASE 1: DISCOVERY
Trigger: On project load or git pull
- Scan file structure (Greenfield vs. Brownfield)
- Index
package.json and docker-compose.yml
- If Brownfield: Calculate Technical Debt Score
- If Greenfield: Initiate Vision Board Interview
Output Format:
PROJECT_TYPE: [Greenfield | Brownfield]
TECH_STACK: [Detected Stack]
DEBT_SCORE: [0-100] (Brownfield only)
READY_FOR: [PHASE_2_VISION_BOARD]
PHASE 2: VISION BOARD
Trigger: Post-Discovery
Ask 3 targeted questions:
Q1: What is the PRIMARY outcome this feature/project must achieve?
Q2: Who is the end user, and what is their skill level?
Q3: What are the NON-NEGOTIABLE constraints (timeline, tech, budget)?
PHASE 3: BLUEPRINT
Trigger: Post-Vision Approval
- Generate
docs/phases/phase-X-spec.md
- Generate/Update
ARCHITECTURE.md
- Output:
PLAN_LOCKED. READY FOR EXECUTION.
PHASE 4: EXECUTION CHUNKS
Execute sequentially. Do NOT proceed to Step B until Step A is confirmed.
| Section | Focus | Verification |
|---|
| SECTION_A | Core Configuration (tsconfig, next.config, env) | pnpm turbo run type-check |
| SECTION_B | Database & Auth Layer | pnpm run docker:up && verify |
| SECTION_C | Backend Logic (API Routes, Server Actions) | cd apps/backend && uv run pytest |
| SECTION_D | Frontend Shell (Layouts, CSS, Design System) | Visual inspection + Lighthouse |
| SECTION_E | Feature Implementation | Full test suite |
Commit After Each Section:
git add . && git commit -m "feat(section-X): [description]"
Response Format
When activating this skill, format responses as:
[AGENT_ACTIVATED]: {agent_name}
[PHASE]: {current_phase}
[SECTION]: {current_section} (if in execution)
[STATUS]: {in_progress | awaiting_verification | complete}
{response_content}
[NEXT_ACTION]: {what happens next}
Verification Gates
Before advancing phases, run:
- Type Check:
pnpm turbo run type-check
- Lint:
pnpm turbo run lint
- Test:
pnpm turbo run test
- Build:
pnpm build (for deployment phases)
Australian Localisation (en-AU)
- Date Format: DD/MM/YYYY
- Time Format: H:MM am/pm (AEST/AEDT)
- Currency: AUD ($)
- Spelling: colour, behaviour, optimisation, analyse, centre
- Tone: Direct, professional, no unnecessary superlatives
Emergency Protocols
Context Overflow Prevention
If approaching context limits:
- Summarise current state
- Commit all work in progress
- Output:
CONTEXT_CHECKPOINT. Resume with: "Continue Phase X, Section Y"
Rollback Protocol
If verification fails:
git stash
git checkout HEAD~1
git stash pop
Quick Commands
pnpm run verify
pnpm dev
pnpm turbo run type-check lint
pnpm turbo run test
Anti-Patterns
| Pattern | Problem | Correct Approach |
|---|
| Executing all phases in one pass | Context overflow, unverified output | Sectional execution with verification gates between each phase |
| Skipping verification gates | Broken code propagates to later phases | Run type-check, lint, and test before advancing |
| No spec before implementation | Scope creep, misaligned deliverables | Generate phase-X-spec.md in Blueprint phase before any code |
| Ignoring Council of Logic checks | Sub-optimal algorithms and architecture | Activate MATH_COUNCIL for complexity and architecture review |
| Committing multiple sections at once | Difficult rollback, unclear git history | Commit after each section with descriptive message |
Checklist
GENESIS PROTOCOL ACTIVE. AWAITING DIRECTIVE.
