NoahDuongMaster/vibe-code-stack-for-ceos
Stop prompting. Start delegating. The full-stack monorepo where Claude Code, Cursor, Gemini CLI, Kiro, Copilot, and Windsurf all read one company handbook — and
AI-first Turborepo monorepo where Claude Code, Cursor, Copilot, Gemini CLI, Kiro & Windsurf read one AGENTS.md handbook and ship identical, production-grade code. Next.js 16 dapp, React 19 admin SPA, Astro landing, Connect-RPC backend — TypeScript strict, deployed to Cloudflare.
vibe-code-stack-for-ceos.duongnamtruong.com
- TypeScript66.2%
- Shell25.3%
- HCL3.8%
- Makefile2.2%
- Dockerfile1.1%
- Astro1.1%
- CSS0.2%
- JavaScript0.1%
1 Review
Vibe Code Stack for CEOs is a serious, carefully engineered starter project for teams that want AI coding agents to follow consistent rules. Its best feature is the detailed AGENTS.md handbook, which defines architecture, naming, testing, security, and completion standards in one place. Those rules are backed by real enforcement through TypeScript checks, Biome, ESLint, Vitest, Playwright, Docker tests, dependency audits, and CI workflows. The repository also includes three frontend apps, three backend services, typed Protobuf contracts, PostgreSQL, Cloudflare infrastructure, Terraform, monitoring, and automated dependency updates. The English and Vietnamese setup guides are unusually thorough, and the use of locked tools and repeatable commands should reduce differences between developer machines.
The main weakness is the gap between the project’s simple CEO-focused pitch and its actual complexity. Running the complete stack requires knowledge of Cloudflare VPC Services, tunnel credentials, AWS infrastructure, Docker, RPC, Terraform, and many environment variables. That makes this more useful for experienced engineering teams than for nontechnical CEOs. The README also says the stack uses CI-gated staging and production deployments, but the current workflow only deploys the landing site. Full-stack deployment exists in the repository but is disabled. That limitation appears in the setup guide, though it should be stated more clearly near the main deployment claims. Claims such as producing identical production-grade code on the first try and being proven by 26 AI agents are interesting, but the repository does not provide a repeatable benchmark or full audit report that outsiders can independently check. Overall, this is an impressive and well-documented engineering template. A simpler starter path, clearer deployment status, and reproducible evidence for its AI performance claims would make it more accessible and credible.
