convez.ai
Skill
chrishorne74

enterprise-architecture-team

chrishorne74/enterprise-architecture-team

Ver no GitHub
InicialPluginScore · de 0 a 100

Claude Agent Skill (+ GitHub Copilot prompt) for setting up and running an enterprise architecture function: charter, vision, principles, standards & governance, and stakeholder engagement. TOGAF-aligned, vendor-neutral.

Instalação

$npx convez-ai chrishorne74/enterprise-architecture-team

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
1
Forks
0
Acompanhando
0
Saúde
42%
Último commit
há 4 meses
Versão
v1.1.0
Issues abertas
0
Contribuidores
1

README

A tradução para o português deste README ainda está na fila.

Enterprise Architecture Team

A Claude skill (with a matching GitHub Copilot prompt-file edition) for establishing, articulating, and running an enterprise architecture (EA) function — the team that aligns technology to business strategy and sets the guardrails everyone else builds within.

It is TOGAF-aligned but vendor-neutral: it borrows TOGAF's structure (the ADM, the principles template, stakeholder management) without forcing heavyweight process onto an organization that doesn't want it.

What it does

  • Charter the team — responsibilities, mandate & decision rights, RACI, governance, operating cadence.
  • Vision — a short, outcome-led target-state statement a CIO would say out loud.
  • Principles — an architecture principles catalog in the four-part form (name, statement, rationale, implications) across business, data, application, technology, and security.
  • Standards & governance — a standards register with a lifecycle (mandated / preferred / contained / retired), a technology reference model / tech radar, the Architecture Review Board, and the exception/waiver process.
  • Stakeholder engagement — who the stakeholders are, what each actually cares about, a power/interest grid, and a per-group engagement plan.

The skill routes to the right depth — a quick explanation, a single artifact, or the whole operating blueprint assembled under one executive summary.

Structure

enterprise-architecture-team/
├── SKILL.md                                  # workflow, mode-routing, grounding guidance
├── references/
│   ├── responsibilities-and-vision.md        # responsibility catalog, team shapes, mandate, maturity, vision
│   ├── principles-and-standards.md           # principle anatomy + quality bar, starter set, standards, ARB, exceptions
│   └── stakeholder-engagement.md             # stakeholder catalog + concerns, power/interest grid, anti-patterns
└── assets/
    ├── ea-charter-template.md                # charter skeleton (mandate, RACI, cadence, metrics)
    ├── architecture-principles-template.md   # principles catalog skeleton (four-part form)
    ├── standards-and-governance-template.md  # standards register, tech radar, ARB ToR, exception process
    └── stakeholder-engagement-template.md    # stakeholder catalog, grid, engagement plan

Usage

Claude (Agent Skills)

Place the enterprise-architecture-team/ folder where your Claude skills are discovered, then ask in natural language — e.g. "help me set up our enterprise architecture team's charter and principles" or "who are the stakeholders for enterprise architecture and what do they care about?" The skill auto-triggers from its description and routes to the right depth.

GitHub Copilot (prompt file)

The same workflow is available to Copilot as a prompt file at .github/prompts/enterprise-architecture-team.prompt.md. In VS Code, Visual Studio, or a JetBrains IDE with Copilot Chat, invoke it on demand with /enterprise-architecture-team. It reuses the same enterprise-architecture-team/references/* and assets/* files, so the two editions stay in sync from one source of truth.

Grounding

Named-framework specifics evolve (TOGAF editions, ArchiMate, the EA tooling landscape). The skill attributes named frameworks (TOGAF, Zachman, ArchiMate, Gartner, capability models) accurately, verifies current wording against official sources when it matters, and flags uncertainty rather than inventing citations. The reference material stands on its own without them.

License

MIT © 2026 Chris Horne

Related

See also the companion cloud-operating-model skill for designing and transitioning cloud operating models.

Ver o original no GitHub

Bom para

Relacionados