Triggers
- NEXUS strategy
- multi-agent orchestration
- agency strategy
- agent coordination
- project pipeline
- NEXUS pipeline
- agent roster
- deployment mode
- NEXUS full
- NEXUS sprint
- NEXUS micro
- coordination matrix
- agent division
- quality gates
- risk management framework
- pipeline status
- project phases
- agency playbook
Instructions
NEXUS (Network of EXperts, Unified in Strategy) transforms independent AI specialists into a synchronized intelligence network. This is the master strategy document governing all phases, agent coordination, handoff protocols, and quality gates.
What NEXUS Solves
Individual agents are powerful but without coordination produce conflicting decisions, duplicated effort, quality gaps at handoffs, and no shared context. NEXUS defines WHO activates at each phase, WHAT they deliver, HOW work flows between agents, and WHEN quality gates trigger.
Deployment Modes
NEXUS-Full (New product, 8-16 weeks):
- All 7 phases (0-6)
- 30+ agents across all divisions
- Complete quality gate system
- Full handoff protocol
NEXUS-Sprint (Major feature, 4-8 weeks):
- Phases 1-5 (skip deep discovery)
- 15-25 agents
- Compressed timelines
- Existing infrastructure leveraged
NEXUS-Micro (Quick fix or campaign, 1-2 weeks):
- Phases 3-5 only (build, verify, deploy)
- 5-10 agents
- Minimal ceremony
- Fastest time-to-delivery
The Seven Phases
Use the corresponding skill for each phase:
- Phase 0 — Intelligence & Discovery (
agency-phase-0-discovery): Validate opportunity. 3-7 days, 6 agents. - Phase 1 — Strategy & Architecture (
agency-phase-1-strategy): Define what to build. 5-10 days, 8 agents. - Phase 2 — Foundation & Scaffolding (
agency-phase-2-foundation): Build technical foundation. 3-5 days, 6 agents. - Phase 3 — Build & Iterate (
agency-phase-3-build): Implement features via Dev-QA loops. 2-12 weeks, 15-30+ agents. - Phase 4 — Quality & Hardening (
agency-phase-4-hardening): Prove production readiness. 3-7 days, 8 agents. - Phase 5 — Launch & Growth (
agency-phase-5-launch): Go-to-market execution. 2-4 weeks, 12 agents. - Phase 6 — Operate & Evolve (
agency-phase-6-operate): Sustained operations. Ongoing, 12+ agents.
Agent Division Structure
Engineering Division: Frontend Developer, Backend Architect, AI Engineer, DevOps Automator, Infrastructure Maintainer, Mobile App Builder, Senior Developer, Rapid Prototyper, XR Developers, Terminal Integration Specialist, LSP/Index Engineer
Design Division: UX Architect, UI Designer, Brand Guardian, Visual Storyteller, Whimsy Injector
Testing Division: Evidence Collector, API Tester, Performance Benchmarker, Test Results Analyzer
Product Division: Studio Producer, Sprint Prioritizer, Senior Project Manager, Project Shepherd, Agents Orchestrator
Research Division: Trend Researcher, UX Researcher, Feedback Synthesizer, Tool Evaluator
Growth Division: Growth Hacker, Content Creator, Social Media Strategist, Twitter Engager, TikTok Strategist, Instagram Curator, Reddit Community Builder, App Store Optimizer
Support Division: Support Responder, Executive Summary Generator, Finance Tracker, Legal Compliance Checker, Workflow Optimizer, Experiment Tracker, Analytics Reporter, Data Analytics Reporter, Studio Operations
Core Mechanics
Dev-QA Loop: Developer implements -> Evidence Collector tests -> PASS/FAIL -> max 3 retries -> escalate. This is the heartbeat of Phase 3.
Quality Gates: Every phase boundary requires evidence-based gate passage. No advancing without proof. Gate keepers vary by phase.
Handoff Protocol: Use agent-handoff skill templates. Consistent handoffs prevent context loss.
Agent Activation: Use agent-activation skill templates. Copy, customize placeholders, deploy via organization_spawn or organization_delegate.
Coordination Matrix
| Phase | Primary Coordinator | Gate Keeper |
|---|---|---|
| Phase 0 | Agents Orchestrator | Executive Summary Generator |
| Phase 1 | Studio Producer | Studio Producer + Reality Checker |
| Phase 2 | DevOps Automator | DevOps Automator + Evidence Collector |
| Phase 3 | Agents Orchestrator | Agents Orchestrator |
| Phase 4 | Reality Checker | Reality Checker (sole authority) |
| Phase 5 | Project Shepherd | Studio Producer + Analytics Reporter |
| Phase 6 | Studio Producer | Studio Producer |
Risk Management Framework
| Risk Category | Detection | Response |
|---|---|---|
| Technical debt | Tool Evaluator quarterly review | Sprint Prioritizer allocates 20% capacity |
| Scope creep | Sprint Prioritizer MoSCoW enforcement | Project Shepherd change management |
| Quality regression | Evidence Collector + Performance Benchmarker | Agents Orchestrator escalation |
| Security vulnerability | Legal Compliance Checker + DevOps Automator | Incident response protocol |
| Market shift | Trend Researcher monthly intelligence | Studio Producer strategic pivot |
Quick-Start Guide
- Determine deployment mode (Full/Sprint/Micro)
- Activate Agents Orchestrator with project spec using
agent-activationskill - Follow phase sequence starting from appropriate phase
- Use handoff templates (
agent-handoffskill) at every boundary - Enforce quality gates — no exceptions
Use knowledge_write to persist the project's NEXUS configuration, agent roster, and coordination decisions.
Deliverables
- Deployment mode selected (Full/Sprint/Micro)
- Agent roster defined for the project
- Phase sequence planned
- Coordination matrix established
- Quality gate criteria defined per phase
- Risk management framework active
- Handoff protocol in use
Success Metrics
- All phases complete with quality gates passed
- Zero context loss at handoff boundaries
- Dev-QA first-pass rate improving sprint over sprint
- All agents activated with complete context
- Quality gate criteria met with evidence
- Risk detection and response within SLA
- Deployment mode appropriate for project scope
Verify
- The deliverable for this phase exists as a concrete artifact (doc, ticket, board, repo) and its location is shared, not described
- Each commitment has an owner name, a due date, and a definition-of-done that someone other than the author could check
- Risks are listed with likelihood/impact and a named mitigation, not as a generic 'risks: TBD' bullet
- Dependencies on other teams/vendors/agents are explicit; an ack from each dependency is recorded or marked 'pending'
- Success criteria for the next phase are numeric or otherwise objectively testable
- A rollback / kill-switch / 'we will stop if X' criterion is written down before work starts