メニュー
Run the complete SDLC pipeline from BRD to deployed application. Orchestrates spec writing, architecture, UI design, implementation, code review, E2E testing, deployment, and PR creation.
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
SOC 職業分類に基づく
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/cwijayasundara/claude_code_forge_v2 --skill buildコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
SKILL.md を表示中
Design the system architecture including layered dependencies, API contracts (endpoints, schemas, errors), data models, folder structure, and deployment topology. Output a detailed design document to `specs/design/` with all decisions justified.
Story-driven code generation with strict quality principles. Use when implementing user stories, writing production code, creating unit tests, or generating code from specs. Enforces small modules, static typing, 50-line function limit, and 100% meaningful test coverage.
Set up local development with Docker Compose, hot reload, health checks, and database migrations. Generate docker-compose.yml, Dockerfiles, .env.example, and document the deployment topology.
| name | build |
| description | Run the complete SDLC pipeline from BRD to deployed application. Orchestrates spec writing, architecture, UI design, implementation, code review, E2E testing, deployment, and PR creation. |
| disable-model-invocation | true |
| argument-hint | [path-to-BRD] |
| context | fork |
/build [path-to-BRD]
| Phase | Agent | Input | Output | Gate |
|---|---|---|---|---|
| 1. Spec | spec-writer | BRD | specs/stories/ | All stories have acceptance criteria |
| 2. Architecture | architect | specs | specs/design/ | Typed schemas, folders, migrations |
| 3. UI Design | ui-designer | specs + arch | specs/design/mockups/ | Every UI story has a mockup |
| 4. Implementation | implementer + agent team | specs + arch + design | backend/, frontend/ | Tests pass, 100% coverage |
| 5. Code Review | code-reviewer + security-reviewer | code | specs/reviews/ | Zero BLOCK findings |
| 6. E2E Testing | test-engineer | specs + code | specs/test_artefacts/ | Playwright tests pass |
| 7. Deployment | architect | arch + code | docker-compose.yml, CI | docker compose up healthy |
| 8. Commit & PR | — | all changes | PR URL | CI passes |
Phase 4 uses agent teams for parallel execution. See /implement for details.
For error recovery procedures, see references/error-recovery.md.
After the human approves Phase 1 (specs) and Phases 2-3 (architecture + UI), the remaining phases
run autonomously via the /auto ratcheting loop. The /build orchestrator:
/auto which reads .claude/program.md each iteration.After human approval of specs + design:
specs/state/ files (iteration-log, learned-rules, failures)..claude/program.md Current Focus with story count and starting status./auto — the autonomous loop handles implementation, review, testing, and deployment./auto stops when all stories pass OR a stopping criterion is hit./build resumes for Phase 8 (commit + PR) or escalates to human.specs/state/learned-rules.md — never deleted.git add -A can stage secrets or build artifacts. Always stage specific directories.