playbooks

home / skills / chachamaru127 / claude-code-harness / setup

setup skill

/skills/setup

This skill unifies project initialization, tool configuration, and 2-Agent setup to accelerate onboarding and ensure consistent workflows.

npx playbooks add skill chachamaru127/claude-code-harness --skill setup

Review the files below or copy the command above to add this skill to your agents.

Files (19)
SKILL.md
4.0 KB
---
name: setup
description: "Unified setup hub: project init, tool setup, 2-agent config, harness-mem, codex CLI, and rule localization. Use when user mentions setup, initialization, new projects, workflow files, CI setup, LSP setup, MCP setup, codex setup, opencode setup, 2-Agent setup, PM coordination, Cursor setup, harness-mem, claude-mem integration, cross-session memory, localize rules, adapt rules. Do NOT load for: implementation work, reviews, build verification, or deployments."
description-en: "Unified setup hub: project init, tool setup, 2-agent config, harness-mem, codex CLI, and rule localization. Use when user mentions setup, initialization, new projects, workflow files, CI setup, LSP setup, MCP setup, codex setup, opencode setup, 2-Agent setup, PM coordination, Cursor setup, harness-mem, claude-mem integration, cross-session memory, localize rules, adapt rules. Do NOT load for: implementation work, reviews, build verification, or deployments."
description-ja: "統合セットアップハブ: プロジェクト初期化、ツール設定、2エージェント構成、harness-mem、Codex CLI、ルールローカライズ。Use when user mentions setup, initialization, new projects, workflow files, CI setup, LSP setup, MCP setup, codex setup, opencode setup, 2-Agent setup, PM coordination, Cursor setup, harness-mem, claude-mem integration, cross-session memory, localize rules, adapt rules. Do NOT load for: implementation work, reviews, build verification, or deployments."
allowed-tools: ["Read", "Write", "Edit", "Grep", "Glob", "Bash"]
argument-hint: "[ci|dev-tools|lsp|mcp|opencode|codex|webhook|remotion|skills|2agent|harness-mem|localize-rules]"
---

# Setup Skills

プロジェクトセットアップ、ツール設定、ワークフロー構成を一元管理するスキル群です。

## ルーティングテーブル

ユーザーの意図に応じて適切な reference にルーティングします。

### プロジェクト初期化

| 機能 | 詳細 |
|------|------|
| **適応的セットアップ** | See [references/adaptive-setup.md](references/adaptive-setup.md) |
| **スキャフォールディング** | See [references/project-scaffolding.md](references/project-scaffolding.md) |
| **ワークフローファイル** | See [references/workflow-files.md](references/workflow-files.md) |
| **設定ファイル** | See [references/claude-settings.md](references/claude-settings.md) |
| **プロジェクト種別確認** | See [references/project-type-detection.md](references/project-type-detection.md) |

### ツールセットアップ

| サブコマンド | 詳細 |
|------------|------|
| **CI/CD** | See [references/ci-setup.md](references/ci-setup.md) |
| **開発ツール (AST-Grep + LSP)** | See [references/dev-tools-setup.md](references/dev-tools-setup.md) |
| **LSP** | See [references/lsp-setup.md](references/lsp-setup.md) |
| **MCP サーバー** | See [references/mcp-setup.md](references/mcp-setup.md) |
| **OpenCode.ai** | See [references/opencode-setup.md](references/opencode-setup.md) |
| **Codex CLI** | See [references/codex-setup.md](references/codex-setup.md) |
| **Webhook** | See [references/webhook-setup.md](references/webhook-setup.md) |
| **Remotion** | See [references/remotion-setup.md](references/remotion-setup.md) |

### 2-Agent ワークフロー

| 機能 | 詳細 |
|------|------|
| **初期設定** | See [references/2agent-setup.md](references/2agent-setup.md) |
| **ファイル更新** | See [references/2agent-updating.md](references/2agent-updating.md) |
| **Cursor連携** | See [references/cursor-setup.md](references/cursor-setup.md) |

### メモリ・ルール

| 機能 | 詳細 |
|------|------|
| **Harness-Mem セットアップ** | See [references/harness-mem.md](references/harness-mem.md) |
| **ルールローカライズ** | See [references/localize-rules.md](references/localize-rules.md) |

## 実行手順

1. ユーザーのリクエストを分類
2. 上記の「ルーティングテーブル」から適切な参照ファイルを読む
3. その内容に従って実行

Overview

This skill is a unified setup hub for initializing projects and configuring developer tooling across the workflow. It centralizes project scaffolding, CI/LSP/MCP/Codex/OpenCode integrations, 2-Agent orchestration, and memory/rule localization so teams can start consistently. Use it to standardize new-project bootstraps and cross-toolchain setup without diving into implementation or review tasks.

How this skill works

The skill classifies the user intent (project init, tool setup, 2-Agent config, memory, or rule localization) and routes to the appropriate setup path. For each route it provides actionable steps to create scaffolding, install and configure CI, LSP, MCP, Codex CLI, OpenCode integrations, or Cursor/2-Agent flows. It also outlines harness-memory and rule-localization procedures for cross-session memory and localized rule behavior. The skill is orchestration-focused and stops before implementation, code review, or deployment tasks.

When to use it

  • Starting a new repository and needing consistent scaffolding and config files
  • Setting up CI/CD pipelines, workflow files, or developer toolchains (LSP, AST-Grep)
  • Configuring Codex CLI, OpenCode, MCP server, webhooks, or other integrations
  • Preparing a 2-Agent workflow or integrating Cursor for agent collaboration
  • Establishing harness-memory, cross-session memory, or localizing agent rules
  • Coordinating project-level setup tasks across PMs and engineers

Best practices

  • Define project type and target runtime first to choose correct scaffolding and presets
  • Keep setup commands idempotent so repeated runs do not break environments
  • Version control workflow and CI templates to make onboarding reproducible
  • Document environment variables and credential steps for tooling like Codex and MCP
  • Localize rules incrementally: test a narrow rule set before broad localization
  • Separate orchestration (this skill) from implementation work to avoid scope creep

Example use cases

  • Initialize a JS project with standardized scaffolding, CI workflow, and LSP config
  • Add Codex CLI and OpenCode integration to an existing repo without touching business code
  • Configure a 2-Agent collaboration flow and hook Cursor for file handoff
  • Set up an MCP server and webhook endpoints for dev workflows
  • Enable harness-memory and localize behavior rules for a multilingual agent deployment

FAQ

Will this skill write code or implement features?

No. The skill focuses on orchestration and configuration guidance. It creates setup steps and files but does not perform implementation, reviews, or deployments.

Can it run repeatedly without side effects?

Yes—best practice is idempotent setup. The skill recommends commands and templates that can be reapplied safely.

How does it handle rule localization and memory?

It outlines procedures to initialize harness-memory, configure cross-session memory stores, and adapt rules for local environments with incremental testing.