claude-gpt-workflow MCP — Casos de uso, Instalar & Demo ao vivo

Por que usar

Principais recursos

Role separation: Claude strategizes, Codex executes
Adversarial plan review between two models
Three feedback loops (A: plan, B: review/fix, C: batch)
Shell-based integration via ask_codex.sh
Quality gate on each iteration

Demo ao vivo

Como fica na prática

claude-gpt-workflow-skill.replay ▶ pronto

0/0

Instalar

Escolha seu cliente

~/Library/Application Support/Claude/claude_desktop_config.json · Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "claude-gpt-workflow-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/longranger2/claude-gpt-workflow",
        "~/.claude/skills/claude-gpt-workflow"
      ],
      "_inferred": true
    }
  }
}

Abra Claude Desktop → Settings → Developer → Edit Config. Reinicie após salvar.

~/.cursor/mcp.json · .cursor/mcp.json

{
  "mcpServers": {
    "claude-gpt-workflow-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/longranger2/claude-gpt-workflow",
        "~/.claude/skills/claude-gpt-workflow"
      ],
      "_inferred": true
    }
  }
}

Cursor usa o mesmo esquema mcpServers que o Claude Desktop. Config de projeto vence a global.

VS Code → Cline → MCP Servers → Edit

{
  "mcpServers": {
    "claude-gpt-workflow-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/longranger2/claude-gpt-workflow",
        "~/.claude/skills/claude-gpt-workflow"
      ],
      "_inferred": true
    }
  }
}

Clique no ícone MCP Servers na barra lateral do Cline, depois "Edit Configuration".

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "claude-gpt-workflow-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/longranger2/claude-gpt-workflow",
        "~/.claude/skills/claude-gpt-workflow"
      ],
      "_inferred": true
    }
  }
}

Mesmo formato do Claude Desktop. Reinicie o Windsurf para aplicar.

~/.continue/config.json

{
  "mcpServers": [
    {
      "name": "claude-gpt-workflow-skill",
      "command": "git",
      "args": [
        "clone",
        "https://github.com/longranger2/claude-gpt-workflow",
        "~/.claude/skills/claude-gpt-workflow"
      ]
    }
  ]
}

O Continue usa um array de objetos de servidor em vez de um map.

~/.config/zed/settings.json

{
  "context_servers": {
    "claude-gpt-workflow-skill": {
      "command": {
        "path": "git",
        "args": [
          "clone",
          "https://github.com/longranger2/claude-gpt-workflow",
          "~/.claude/skills/claude-gpt-workflow"
        ]
      }
    }
  }
}

Adicione em context_servers. Zed recarrega automaticamente ao salvar.

claude mcp add claude-gpt-workflow-skill -- git clone https://github.com/longranger2/claude-gpt-workflow ~/.claude/skills/claude-gpt-workflow

Uma linha só. Verifique com claude mcp list. Remova com claude mcp remove.

Casos de uso

Usos do mundo real: claude-gpt-workflow

Stress-test a plan by having another model attack it

👤 Engineers doing high-stakes refactors ⏱ ~45 min intermediate

Quando usar: Before touching code on a migration or risky refactor, you want a second model to poke holes.

Pré-requisitos

Codex CLI installed and authenticated — Install OpenAI Codex CLI with your API key
Skill installed — git clone https://github.com/longranger2/claude-gpt-workflow ~/.claude/skills/claude-gpt-workflow

Fluxo

Draft a plan with Claude

Write a plan.md for migrating our auth module from JWT to session cookies.✓ Copiado

→ Step-by-step plan written to plan.md
Trigger adversarial review

/plan-review plan.md✓ Copiado

→ Codex returns critique with risks Claude missed
Iterate until both agree

Incorporate Codex's critique, re-submit, repeat until clean.✓ Copiado

→ Plan reaches stable state

Resultado: A plan that has survived real adversarial review, not just self-grading.

Armadilhas

Endless review loops — Cap at 3 rounds — diminishing returns after

Combine com: filesystem · git

Delegate a batch refactor to Codex with Claude reviewing

👤 Teams doing mechanical multi-file changes ⏱ ~60 min intermediate

Quando usar: A rename / signature change across 50+ files.

Fluxo

Plan and approve

Plan + review a rename of <old> to <new> across the codebase.✓ Copiado

→ Approved plan.md
Execute

/plan-execute plan.md — Codex does the edits, Claude reviews each batch.✓ Copiado

→ Files change in batches, each passes Claude's review

Resultado: A clean multi-file change with two sets of eyes on every batch.

Armadilhas

Codex does something Claude would have rejected — The /plan-execute loop catches it — but only if quality criteria in the plan are explicit

Combine com: git

Implement a feature with iterative quality refinement

👤 Devs who've seen one-shot AI code fall over on edge cases ⏱ ~90 min advanced

Quando usar: A feature complex enough that a single generation won't get it right.

Fluxo

Write plan with acceptance criteria

Plan a feature with explicit acceptance tests, then /plan-execute.✓ Copiado

→ Plan with testable criteria
Watch the loop

Let the review/fix loop run — summarize each iteration.✓ Copiado

→ Iterations converge to passing criteria

Resultado: A feature that passes explicit acceptance, not just a commit that compiles.

Combinações

Combine com outros MCPs para 10× de alavancagem

claude-gpt-workflow-skill + git

Commit after each loop iteration so you can revert if the loop diverges

Before each /plan-execute iteration, create a git tag.✓ Copiado

claude-gpt-workflow-skill + filesystem

Plan and critiques live in the repo

Write plan.md and review-notes.md to docs/ and keep them in git.✓ Copiado

Ferramentas

O que este MCP expõe

Ferramenta	Entradas	Quando chamar	Custo
ask_codex.sh	request string	Delegate a coding task	Codex API tokens
/plan-review	plan file path	Before execution	Codex API tokens
/plan-execute	approved plan path	After /plan-review passes	Codex tokens per iteration

Custo e limites

O que custa rodar

Cota de API: Codex API tokens — potentially 2–5x a single-model workflow since both models run
Tokens por chamada: High — plan + review + execute
Monetário: Two API subscriptions (Claude + OpenAI)
Dica: Reserve this for risky work; don't use it for small one-file edits

Segurança

Permissões, segredos, alcance

Armazenamento de credenciais: OPENAI_API_KEY in env for Codex CLI

Saída de dados: Code leaves your machine to both providers — check corporate policy

Two-model workflow doubles the surface that sees your source.

Solução de problemas

Erros comuns e correções

ask_codex.sh: command not found

Codex CLI not on PATH or not installed

Verificar: which codex

Plan-execute loops forever

Acceptance criteria too vague — tighten the plan or cap iterations

Models disagree and plan never stabilizes

Force a tiebreaker round with a shorter scope

Alternativas

claude-gpt-workflow vs. outros

Alternativa	Quando usar	Troca
Single-model Claude Code	The task isn't risky enough to justify two-model overhead	Fewer adversarial signals
agent-skill-tdd	Discipline comes from tests, not from another model's critique	No cross-model review

Mais

Recursos

📖 Leia o README oficial no GitHub

🐙 Ver issues abertas

🔍 Ver todos os 400+ servidores MCP e Skills