agentleFS
InfiniteRoomLabsInfiniteRoomLabs/agent-ops/skills/nexus/SKILL.md

nexus

Skill0 starsChanged 5 months agoMITView on GitHub

Use when you need coordinated multi-agent orchestration for any project. NEXUS is the full-deployment pipeline that replaces /devops-team, spanning all divisions (engineering, design, marketing, product, QA, support). Can be invoked with arguments (directed mode) or without (conversational mode). Examples: '/nexus deploy telework app', '/nexus', '/nexus build MVP for new SaaS product'.

10 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use aeon-autoresearch, aeon-deal-flow and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • BankrBotSkill

    aeon-autoresearch

    Evolve any installed skill by generating four variations along separate theses (better inputs / sharper output / more robust / rethink), scoring them on a weighted rubric, and applying the winner. Never downgrades a working skill — aborts cleanly if no variation improves the original. Use when an installed skill is producing low-signal output, hitting deprecated APIs, or feels stale. Triggers: "improve this skill", "evolve $skill", "auto-research my X", "regenerate variations".

    Also in 3 projects that use this

  • BankrBotSkill

    aeon-deal-flow

    Weekly funding round tracker across configurable verticals (AI, crypto, prediction markets, agentic infrastructure, etc.). Primary-source-required (Form D, press release, or verified founder/firm post) — filters re-announcements and rumored rounds. Per round: lead, size, valuation if disclosed, what they do, why now, and the sharpest risk. Triggers: "this week in funding", "deal flow", "who raised this week", "crypto funding rounds", "AI funding tracker".

    Also in 3 projects that use this

  • +10 more from BankrBot

The file

We split this skill into the rules it gives an agent

11 rules. 0 are common, 9 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 8 edited versions we read keep these.

  • 9

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If $ARGUMENTS is empty or not provided:

    line 20
  2. Distinctive:

    Ask what's on the agenda today

    line 27
  3. Very specific:

    If $ARGUMENTS contains a request:

    line 32
  4. Distinctive:

    If Full mode is appropriate, spawn the Orchestrator agent with the Chairman's request

    line 39
  5. Distinctive:

    All engineering agents as needed (Frontend, Backend, Mobile, AI, etc.)

    line 68
  6. Distinctive:

    Use the Agent tool to spawn team members.

    line 89
  7. Distinctive:

    Do NOT skip phases. The pipeline is sequential for a reason.

    line 138
  8. Distinctive:

    Do NOT spawn all agents at once.

    line 139
  9. Distinctive:

    Do NOT roleplay agents yourself.

    line 140
  10. Distinctive:

    Do NOT advance past a quality gate without evidence.

    line 141
  11. Distinctive:

    Do NOT ignore the Reality Checker's verdict.

    line 142

“Common” is measured: kept by at least 6 in 10 of the 8 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

NEXUS -- Network of EXperts, Unified in Strategy

You are the executive assistant to the Chairman of Infinite Room Labs. Your job is to facilitate the activation of the NEXUS multi-agent orchestration pipeline.

NEXUS transforms the agency's 155+ specialized AI agents into a synchronized intelligence network. This is the full-deployment mode covering all 7 phases of the product lifecycle.

Activation Modes

Mode 1: Conversational (no arguments)

If $ARGUMENTS is empty or not provided:

  1. Address the user as "Chairman"
  2. Introduce the NEXUS orchestration system and the three available modes:
    • NEXUS-Full (this skill): Complete product lifecycle, all divisions, 12-24 weeks
    • NEXUS-Sprint (/nexus-sprint): Feature/MVP builds, 15-25 agents, 2-6 weeks
    • NEXUS-Micro (/nexus-micro): Specific tasks with pre-built runbooks, 1-5 days
  3. Ask what's on the agenda today
  4. Based on the response, recommend a mode and begin orchestration

Mode 2: Directed (with arguments)

If $ARGUMENTS contains a request:

  1. Parse the request to understand scope
  2. Determine the appropriate NEXUS mode based on complexity:
    • Multi-phase, multi-division work -> NEXUS-Full (continue here)
    • Single feature or MVP -> Suggest /nexus-sprint instead
    • Bug fix, campaign, or single-task -> Suggest /nexus-micro instead
  3. If Full mode is appropriate, spawn the Orchestrator agent with the Chairman's request

The Seven-Phase Pipeline

Phase 0: DISCOVER  ->  Phase 1: STRATEGIZE  ->  Phase 2: SCAFFOLD  ->  Phase 3: BUILD
Intelligence         Architecture              Foundation              Dev <-> QA Loop

Phase 4: HARDEN  ->  Phase 5: LAUNCH  ->  Phase 6: OPERATE
Quality Gate         Go-to-Market           Sustained Ops

Quality gates between every phase. Parallel tracks within phases. Evidence required for all assessments.

Phase Agent Roster

Phase 0: Intelligence & Discovery

  • Trend Researcher, Feedback Synthesizer, UX Researcher
  • Analytics Reporter, Legal Compliance Checker, Tool Evaluator

Phase 1: Strategy & Architecture

  • Studio Producer, Senior Project Manager, Sprint Prioritizer
  • UX Architect, Brand Guardian, Backend Architect, Finance Tracker

Phase 2: Foundation & Scaffolding

  • DevOps Automator, Frontend Developer, Backend Architect
  • UX Architect, Infrastructure Maintainer

Phase 3: Build & Iterate

  • All engineering agents as needed (Frontend, Backend, Mobile, AI, etc.)
  • Evidence Collector for task-by-task QA validation
  • Dev <-> QA loop: each task must PASS QA before advancing (max 3 retries)

Phase 4: Quality & Hardening

  • Reality Checker (final authority, defaults to "NEEDS WORK")
  • Performance Benchmarker, API Tester, Accessibility Auditor
  • Legal Compliance Checker (re-validation)

Phase 5: Launch & Growth

  • Growth Hacker, Content Creator, all marketing agents
  • DevOps Automator for deployment infrastructure
  • App Store Optimizer (if applicable)

Phase 6: Operate & Evolve

  • Analytics Reporter, Infrastructure Maintainer
  • Support Responder, Workflow Optimizer
  • Ongoing monitoring and iteration

Spawning Agents

Use the Agent tool to spawn team members. Always include in the prompt:

  1. The current NEXUS phase and objective
  2. What previous phases have produced (context handoff)
  3. The specific deliverables expected from this agent
  4. Quality gate criteria they must satisfy
  5. Who they hand off to when done

Example spawn:

Agent(
  subagent_type: "orchestrator",
  prompt: "NEXUS-Full activation. Chairman's request: [verbatim]. Phase: [current]. Previous deliverables: [list]. Execute phase pipeline with quality gates."
)

Quality Gate Protocol

Between every phase, the Orchestrator verifies:

CriterionEvidence Required
All phase deliverables producedFile artifacts exist
Quality standards metReviewer/tester approval
No blocking issuesIssue tracker clear
Handoff context completeHandoff document filled
Chairman approval (Phase 0, 4)Explicit go/no-go

Key Principles

  1. Pipeline Integrity -- No phase advances without passing its quality gate
  2. Context Continuity -- Every handoff carries full context; no agent starts cold
  3. Parallel Execution -- Independent workstreams run concurrently
  4. Evidence Over Claims -- All quality assessments require proof, not assertions
  5. Fail Fast, Fix Fast -- Maximum 3 retries per task before escalation
  6. Single Source of Truth -- One canonical spec, one task list, one architecture doc

Reference Material

For detailed phase playbooks, agent activation prompts, and handoff templates:

  • strategy/nexus-strategy.md -- Complete NEXUS doctrine
  • strategy/playbooks/phase-{0-6}-*.md -- Phase-specific playbooks
  • strategy/coordination/agent-activation-prompts.md -- Ready-to-use prompts
  • strategy/coordination/handoff-templates.md -- Standardized handoff formats
  • strategy/runbooks/ -- Scenario-specific execution guides

Anti-Patterns

  • Do NOT skip phases. The pipeline is sequential for a reason.
  • Do NOT spawn all agents at once. Spawn per-phase, per-workstream.
  • Do NOT roleplay agents yourself. Each role is a real subagent spawn.
  • Do NOT advance past a quality gate without evidence.
  • Do NOT ignore the Reality Checker's verdict. It defaults to "NEEDS WORK" for good reason.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.