⚡ NEXUS Quick-Start Guide agent

> Get from zero to orchestrated multi-agent pipeline in 5 minutes.

by msitarzewski·MIT license·★ 154,023 Stars on the repo·GitHub ↗

Files of ⚡ NEXUS Quick-Start Guide

msitarzewski/main1 file
QUICKSTART.md
Show the full text195 lines

⚡ NEXUS Quick-Start Guide

Get from zero to orchestrated multi-agent pipeline in 5 minutes.


What is NEXUS?

NEXUS (Network of EXperts, Unified in Strategy) turns The Agency's AI specialists into a coordinated pipeline. Instead of activating agents one at a time and hoping they work together, NEXUS defines exactly who does what, when, and how quality is verified at every step.

Choose Your Mode

I want to... Use Agents Time
Build a complete product from scratch NEXUS-Full All 12-24 weeks
Build a feature or MVP NEXUS-Sprint 15-25 2-6 weeks
Do a specific task (bug fix, campaign, audit) NEXUS-Micro 5-10 1-5 days

🚀 NEXUS-Full: Start a Complete Project

Copy this prompt to activate the full pipeline:

Activate Agents Orchestrator in NEXUS-Full mode.

Project: [YOUR PROJECT NAME]
Specification: [DESCRIBE YOUR PROJECT OR LINK TO SPEC]

Execute the complete NEXUS pipeline:
- Phase 0: Discovery (Trend Researcher, Feedback Synthesizer, UX Researcher, Analytics Reporter, Legal Compliance Checker, Tool Evaluator)
- Phase 1: Strategy (Studio Producer, Senior Project Manager, Sprint Prioritizer, UX Architect, Brand Guardian, Backend Architect, Finance Tracker)
- Phase 2: Foundation (DevOps Automator, Frontend Developer, Backend Architect, UX Architect, Infrastructure Maintainer)
- Phase 3: Build (Dev↔QA loops — all engineering + Evidence Collector)
- Phase 4: Harden (Reality Checker, Performance Benchmarker, API Tester, Legal Compliance Checker)
- Phase 5: Launch (Growth Hacker, Content Creator, all marketing agents, DevOps Automator)
- Phase 6: Operate (Analytics Reporter, Infrastructure Maintainer, Support Responder, ongoing)

Quality gates between every phase. Evidence required for all assessments.
Maximum 3 retries per task before escalation.

🏃 NEXUS-Sprint: Build a Feature or MVP

Copy this prompt:

Activate Agents Orchestrator in NEXUS-Sprint mode.

Feature/MVP: [DESCRIBE WHAT YOU'RE BUILDING]
Timeline: [TARGET WEEKS]
Skip Phase 0 (market already validated).

Sprint team:
- PM: Senior Project Manager, Sprint Prioritizer
- Design: UX Architect, Brand Guardian
- Engineering: Frontend Developer, Backend Architect, DevOps Automator
- QA: Evidence Collector, Reality Checker, API Tester
- Support: Analytics Reporter

Begin at Phase 1 with architecture and sprint planning.
Run Dev↔QA loops for all implementation tasks.
Reality Checker approval required before launch.

🎯 NEXUS-Micro: Do a Specific Task

Pick your scenario and copy the prompt:

Fix a Bug
Activate Backend Architect to investigate and fix [BUG DESCRIPTION].
After fix, activate API Tester to verify the fix.
Then activate Evidence Collector to confirm no visual regressions.
Run a Marketing Campaign
Activate Social Media Strategist as campaign lead for [CAMPAIGN DESCRIPTION].
Team: Content Creator, Twitter Engager, Instagram Curator, Reddit Community Builder.
Brand Guardian reviews all content before publishing.
Analytics Reporter tracks performance daily.
Growth Hacker optimizes channels weekly.
Conduct a Compliance Audit
Activate Legal Compliance Checker for comprehensive compliance audit.
Scope: [GDPR / CCPA / HIPAA / ALL]
After audit, activate Executive Summary Generator to create stakeholder report.
Investigate Performance Issues
Activate Performance Benchmarker to diagnose performance issues.
Scope: [API response times / Page load / Database queries / All]
After diagnosis, activate Infrastructure Maintainer for optimization.
DevOps Automator deploys any infrastructure changes.
Market Research
Activate Trend Researcher for market intelligence on [DOMAIN].
Deliverables: Competitive landscape, market sizing, trend forecast.
After research, activate Executive Summary Generator for executive brief.
UX Improvement
Activate UX Researcher to identify usability issues in [FEATURE/PRODUCT].
After research, activate UX Architect to design improvements.
Frontend Developer implements changes.
Evidence Collector verifies improvements.

📁 Strategy Documents

Document Purpose Location
Master Strategy Complete NEXUS doctrine strategy/nexus-strategy.md
Phase 0 Playbook Discovery & intelligence strategy/playbooks/phase-0-discovery.md
Phase 1 Playbook Strategy & architecture strategy/playbooks/phase-1-strategy.md
Phase 2 Playbook Foundation & scaffolding strategy/playbooks/phase-2-foundation.md
Phase 3 Playbook Build & iterate strategy/playbooks/phase-3-build.md
Phase 4 Playbook Quality & hardening strategy/playbooks/phase-4-hardening.md
Phase 5 Playbook Launch & growth strategy/playbooks/phase-5-launch.md
Phase 6 Playbook Operate & evolve strategy/playbooks/phase-6-operate.md
Activation Prompts Ready-to-use agent prompts strategy/coordination/agent-activation-prompts.md
Handoff Templates Standardized handoff formats strategy/coordination/handoff-templates.md
Startup MVP Runbook 4-6 week MVP build strategy/runbooks/scenario-startup-mvp.md
Enterprise Feature Runbook Enterprise feature development strategy/runbooks/scenario-enterprise-feature.md
Marketing Campaign Runbook Multi-channel campaign strategy/runbooks/scenario-marketing-campaign.md
Incident Response Runbook Production incident handling strategy/runbooks/scenario-incident-response.md

🔑 Key Concepts in 30 Seconds

  1. Quality Gates — No phase advances without evidence-based approval
  2. Dev↔QA Loop — Every task is built then tested; PASS to proceed, FAIL to retry (max 3)
  3. Handoffs — Structured context transfer between agents (never start cold)
  4. Reality Checker — Final quality authority; defaults to "NEEDS WORK"
  5. Agents Orchestrator — Pipeline controller managing the entire flow
  6. Evidence Over Claims — Screenshots, test results, and data — not assertions

🎭 The Agents at a Glance

ENGINEERING         │ DESIGN              │ MARKETING
Frontend Developer  │ UI Designer         │ Growth Hacker
Backend Architect   │ UX Researcher       │ Content Creator
Mobile App Builder  │ UX Architect        │ Twitter Engager
AI Engineer         │ Brand Guardian      │ TikTok Strategist
DevOps Automator    │ Visual Storyteller  │ Instagram Curator
Rapid Prototyper    │ Whimsy Injector     │ Reddit Community Builder
Senior Developer    │ Image Prompt Eng.   │ App Store Optimizer
                    │                     │ Social Media Strategist
────────────────────┼─────────────────────┼──────────────────────
PRODUCT             │ PROJECT MGMT        │ TESTING
Sprint Prioritizer  │ Studio Producer     │ Evidence Collector
Trend Researcher    │ Project Shepherd    │ Reality Checker
Feedback Synthesizer│ Studio Operations   │ Test Results Analyzer
                    │ Experiment Tracker  │ Performance Benchmarker
                    │ Senior Project Mgr  │ API Tester
                    │                     │ Tool Evaluator
                    │                     │ Workflow Optimizer
────────────────────┼─────────────────────┼──────────────────────
SUPPORT             │ SPATIAL             │ SPECIALIZED
Support Responder   │ XR Interface Arch.  │ Agents Orchestrator
Analytics Reporter  │ macOS Spatial/Metal │ Analytics Reporter
Finance Tracker     │ XR Immersive Dev    │ LSP/Index Engineer
Infra Maintainer    │ XR Cockpit Spec.    │ Sales Data Extraction
Legal Compliance    │ visionOS Spatial    │ Data Consolidation
Exec Summary Gen.   │ Terminal Integration│ Report Distribution

Start with a mode. Follow the playbook. Trust the pipeline.

strategy/nexus-strategy.md — The complete doctrine

1# ⚡ NEXUS Quick-Start Guide
2 
3> **Get from zero to orchestrated multi-agent pipeline in 5 minutes.**
4 
5---
6 
7## What is NEXUS?
8 
9**NEXUS** (Network of EXperts, Unified in Strategy) turns The Agency's AI specialists into a coordinated pipeline. Instead of activating agents one at a time and hoping they work together, NEXUS defines exactly who does what, when, and how quality is verified at every step.
10 
11## Choose Your Mode
12 
13| I want to... | Use | Agents | Time |
14|-------------|-----|--------|------|
15| Build a complete product from scratch | **NEXUS-Full** | All | 12-24 weeks |
16| Build a feature or MVP | **NEXUS-Sprint** | 15-25 | 2-6 weeks |
17| Do a specific task (bug fix, campaign, audit) | **NEXUS-Micro** | 5-10 | 1-5 days |
18 
19---
20 
21## 🚀 NEXUS-Full: Start a Complete Project
22 
23**Copy this prompt to activate the full pipeline:**
24 
25```
26Activate Agents Orchestrator in NEXUS-Full mode.
27 
28Project: [YOUR PROJECT NAME]
29Specification: [DESCRIBE YOUR PROJECT OR LINK TO SPEC]
30 
31Execute the complete NEXUS pipeline:
32- Phase 0: Discovery (Trend Researcher, Feedback Synthesizer, UX Researcher, Analytics Reporter, Legal Compliance Checker, Tool Evaluator)
33- Phase 1: Strategy (Studio Producer, Senior Project Manager, Sprint Prioritizer, UX Architect, Brand Guardian, Backend Architect, Finance Tracker)
34- Phase 2: Foundation (DevOps Automator, Frontend Developer, Backend Architect, UX Architect, Infrastructure Maintainer)
35- Phase 3: Build (Dev↔QA loops — all engineering + Evidence Collector)
36- Phase 4: Harden (Reality Checker, Performance Benchmarker, API Tester, Legal Compliance Checker)
37- Phase 5: Launch (Growth Hacker, Content Creator, all marketing agents, DevOps Automator)
38- Phase 6: Operate (Analytics Reporter, Infrastructure Maintainer, Support Responder, ongoing)
39 
40Quality gates between every phase. Evidence required for all assessments.
41Maximum 3 retries per task before escalation.
42```
43 
44---
45 
46## 🏃 NEXUS-Sprint: Build a Feature or MVP
47 
48**Copy this prompt:**
49 
50```
51Activate Agents Orchestrator in NEXUS-Sprint mode.
52 
53Feature/MVP: [DESCRIBE WHAT YOU'RE BUILDING]
54Timeline: [TARGET WEEKS]
55Skip Phase 0 (market already validated).
56 
57Sprint team:
58- PM: Senior Project Manager, Sprint Prioritizer
59- Design: UX Architect, Brand Guardian
60- Engineering: Frontend Developer, Backend Architect, DevOps Automator
61- QA: Evidence Collector, Reality Checker, API Tester
62- Support: Analytics Reporter
63 
64Begin at Phase 1 with architecture and sprint planning.
65Run Dev↔QA loops for all implementation tasks.
66Reality Checker approval required before launch.
67```
68 
69---
70 
71## 🎯 NEXUS-Micro: Do a Specific Task
72 
73**Pick your scenario and copy the prompt:**
74 
75### Fix a Bug
76```
77Activate Backend Architect to investigate and fix [BUG DESCRIPTION].
78After fix, activate API Tester to verify the fix.
79Then activate Evidence Collector to confirm no visual regressions.
80```
81 
82### Run a Marketing Campaign
83```
84Activate Social Media Strategist as campaign lead for [CAMPAIGN DESCRIPTION].
85Team: Content Creator, Twitter Engager, Instagram Curator, Reddit Community Builder.
86Brand Guardian reviews all content before publishing.
87Analytics Reporter tracks performance daily.
88Growth Hacker optimizes channels weekly.
89```
90 
91### Conduct a Compliance Audit
92```
93Activate Legal Compliance Checker for comprehensive compliance audit.
94Scope: [GDPR / CCPA / HIPAA / ALL]
95After audit, activate Executive Summary Generator to create stakeholder report.
96```
97 
98### Investigate Performance Issues
99```
100Activate Performance Benchmarker to diagnose performance issues.
101Scope: [API response times / Page load / Database queries / All]
102After diagnosis, activate Infrastructure Maintainer for optimization.
103DevOps Automator deploys any infrastructure changes.
104```
105 
106### Market Research
107```
108Activate Trend Researcher for market intelligence on [DOMAIN].
109Deliverables: Competitive landscape, market sizing, trend forecast.
110After research, activate Executive Summary Generator for executive brief.
111```
112 
113### UX Improvement
114```
115Activate UX Researcher to identify usability issues in [FEATURE/PRODUCT].
116After research, activate UX Architect to design improvements.
117Frontend Developer implements changes.
118Evidence Collector verifies improvements.
119```
120 
121---
122 
123## 📁 Strategy Documents
124 
125| Document | Purpose | Location |
126|----------|---------|----------|
127| **Master Strategy** | Complete NEXUS doctrine | `strategy/nexus-strategy.md` |
128| **Phase 0 Playbook** | Discovery & intelligence | `strategy/playbooks/phase-0-discovery.md` |
129| **Phase 1 Playbook** | Strategy & architecture | `strategy/playbooks/phase-1-strategy.md` |
130| **Phase 2 Playbook** | Foundation & scaffolding | `strategy/playbooks/phase-2-foundation.md` |
131| **Phase 3 Playbook** | Build & iterate | `strategy/playbooks/phase-3-build.md` |
132| **Phase 4 Playbook** | Quality & hardening | `strategy/playbooks/phase-4-hardening.md` |
133| **Phase 5 Playbook** | Launch & growth | `strategy/playbooks/phase-5-launch.md` |
134| **Phase 6 Playbook** | Operate & evolve | `strategy/playbooks/phase-6-operate.md` |
135| **Activation Prompts** | Ready-to-use agent prompts | `strategy/coordination/agent-activation-prompts.md` |
136| **Handoff Templates** | Standardized handoff formats | `strategy/coordination/handoff-templates.md` |
137| **Startup MVP Runbook** | 4-6 week MVP build | `strategy/runbooks/scenario-startup-mvp.md` |
138| **Enterprise Feature Runbook** | Enterprise feature development | `strategy/runbooks/scenario-enterprise-feature.md` |
139| **Marketing Campaign Runbook** | Multi-channel campaign | `strategy/runbooks/scenario-marketing-campaign.md` |
140| **Incident Response Runbook** | Production incident handling | `strategy/runbooks/scenario-incident-response.md` |
141 
142---
143 
144## 🔑 Key Concepts in 30 Seconds
145 
1461. **Quality Gates** — No phase advances without evidence-based approval
1472. **Dev↔QA Loop** — Every task is built then tested; PASS to proceed, FAIL to retry (max 3)
1483. **Handoffs** — Structured context transfer between agents (never start cold)
1494. **Reality Checker** — Final quality authority; defaults to "NEEDS WORK"
1505. **Agents Orchestrator** — Pipeline controller managing the entire flow
1516. **Evidence Over Claims** — Screenshots, test results, and data — not assertions
152 
153---
154 
155## 🎭 The Agents at a Glance
156 
157```
158ENGINEERING │ DESIGN │ MARKETING
159Frontend Developer │ UI Designer │ Growth Hacker
160Backend Architect │ UX Researcher │ Content Creator
161Mobile App Builder │ UX Architect │ Twitter Engager
162AI Engineer │ Brand Guardian │ TikTok Strategist
163DevOps Automator │ Visual Storyteller │ Instagram Curator
164Rapid Prototyper │ Whimsy Injector │ Reddit Community Builder
165Senior Developer │ Image Prompt Eng. │ App Store Optimizer
166 │ │ Social Media Strategist
167────────────────────┼─────────────────────┼──────────────────────
168PRODUCT │ PROJECT MGMT │ TESTING
169Sprint Prioritizer │ Studio Producer │ Evidence Collector
170Trend Researcher │ Project Shepherd │ Reality Checker
171Feedback Synthesizer│ Studio Operations │ Test Results Analyzer
172 │ Experiment Tracker │ Performance Benchmarker
173 │ Senior Project Mgr │ API Tester
174 │ │ Tool Evaluator
175 │ │ Workflow Optimizer
176────────────────────┼─────────────────────┼──────────────────────
177SUPPORT │ SPATIAL │ SPECIALIZED
178Support Responder │ XR Interface Arch. │ Agents Orchestrator
179Analytics Reporter │ macOS Spatial/Metal │ Analytics Reporter
180Finance Tracker │ XR Immersive Dev │ LSP/Index Engineer
181Infra Maintainer │ XR Cockpit Spec. │ Sales Data Extraction
182Legal Compliance │ visionOS Spatial │ Data Consolidation
183Exec Summary Gen. │ Terminal Integration│ Report Distribution
184```
185 
186---
187 
188<div align="center">
189 
190**Start with a mode. Follow the playbook. Trust the pipeline.**
191 
192`strategy/nexus-strategy.md` — The complete doctrine
193 
194</div>
195 

Discussion

Alternatives

🎯 NEXUS Agent Activation Prompts> Ready-to-use prompt templates for activating any agent within the NEXUS pipeline. Copy, customize the [PLACEHOLDERS], and deploy.Business & ops · MITArbor — Autonomous Optimization via Hypothesis Tree RefinementAutonomously improve a real artifact (code, training recipe, agent harness, data pipeline, prompt) against an objective and an evaluator, using Hypothesis Tree Refinement (HTR) from the Arbor paper. Use this whenever someone wants to iteratively optimize something over many experiments without overfitting — e.g. "get my model's eval score up", "improve this agent/harness", "tune this pipeline", "beat the baseline on this benchmark", "run a search over approaches and keep the best", "do an MLE-bench / Kaggle-style optimization", or any long-horizon "make this artifact better and don't just memorize the dev set" task. Trigger it even when the user doesn't say "Arbor" or "hypothesis tree" but describes repeated experiment-and-evaluate loops, branching exploration of competing ideas, or worries about a dev/test gap. Runs Claude itself as the coordinator with subagent executors in isolated git worktrees; for the standalone `arbor` CLI tool see references/arbor-upstream.md.Science · MIT/cs:cro-review — CRO Forcing Questions/cs:cro-review <plan> — Pipeline-paranoid interrogation of revenue, win rate, NRR, and ramp time. Use when the forecast misses pipeline coverage, win rates drop, or before scaling the sales team. · MITc-level-agents — Founder-Mode Executive TeamFounder-mode executive team. 13 cs-* C-suite agents (CFO, CMO, CRO, CPO, COO, CHRO, CISO, GC, CDO, CAIO, CCO, VPE, Chief of Staff) and 21 /cs:* slash commands for forcing-question office hours, multi-role boardroom deliberation, strategic sprint pipeline, and meta routing. Use when the founder needs a virtual executive team, when invoking /cs:* commands, or when orchestrating multi-role decisions. · MIT