Zero-Config Code Flow (ZCF) automates code generation and workflows using Claude Code and Codex. Developers and operations teams benefit from streamlined code creation, testing, and deployment. It connects to Node.js and TypeScript projects, integrating with CI/CD pipelines and developer tools.
git clone https://github.com/UfoMiao/zcf.githttps://zcf.ufomiao.com/
1. **Initialize the Project**: Clone your base template (e.g., `git clone https://github.com/your-org/node-ts-starter.git phoenix-api`) and install dependencies (`npm install`). 2. **Configure ZCF**: In your terminal, run: ```bash npx @claude-ai/zcf init --feature="user-authentication" --framework="jest" --lint-tool="eslint" ``` Replace placeholders with your project’s specifics (e.g., `--feature="payment-gateway"`). 3. **Customize Placeholders**: Edit the generated `zcf-config.json` to match your project’s needs (e.g., update `edge_cases` for your domain logic). 4. **Execute the Flow**: Run the ZCF workflow: ```bash npx @claude-ai/zcf generate ``` This triggers Claude Code to auto-generate code, tests, and CI/CD files. 5. **Review and Deploy**: Inspect the `zcf-output/` directory, fix any flagged issues (e.g., flaky tests), and merge into your main branch to trigger the CI/CD pipeline. **Tips for Better Results**: - Use descriptive feature names (e.g., `feature="real-time-notifications"` instead of `feature="notifs"`). - For complex projects, break the workflow into smaller ZCF runs (e.g., `zcf generate --feature="auth"` followed by `zcf generate --feature="payments"`). - Enable verbose logging with `--debug` to troubleshoot generation issues.
Quickly initialize a coding environment with a single command using ZCF.
Easily switch between different language interfaces for international development teams.
Automate the setup of workflows and API integrations for AI-driven projects.
Update existing workflows without manual reconfiguration, saving time and effort.
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/UfoMiao/zcfCopy the install command above and run it in your terminal.
Launch Claude Code, Cursor, or your preferred AI coding agent.
Use the prompt template or examples below to test the skill.
Adapt the skill to your specific use case and workflow.
Generate a Zero-Config Code Flow (ZCF) implementation for [PROJECT_NAME] using Claude Code and Codex. The workflow should: 1) Auto-generate TypeScript/Node.js code for [SPECIFIC_FEATURE], 2) Create unit tests using [TEST_FRAMEWORK], 3) Run linting with [LINT_TOOL], and 4) Output a ready-to-deploy artifact. Include error handling for [EDGE_CASES]. Use [PROJECT_STRUCTURE] as the base template.
Here’s the ZCF-generated implementation for **Project Phoenix**, a Node.js/TypeScript API for user authentication: ### Generated Code Structure ``` phoenix-api/ ├── src/ │ ├── controllers/auth.controller.ts │ ├── services/jwt.service.ts │ ├── models/user.model.ts │ └── utils/errorHandler.ts ├── tests/ │ ├── auth.controller.test.ts │ └── jwt.service.test.ts ├── .eslintrc.json └── package.json ``` ### Key Components 1. **Auth Controller**: Auto-generated REST endpoints for `/login` and `/register` with JWT validation. Includes rate-limiting middleware and input sanitization. 2. **JWT Service**: Type-safe token generation/validation using `jsonwebtoken` and `zod` for schema validation. Handles edge cases like expired tokens and malformed payloads. 3. **Error Handling**: Centralized error handler that logs to Winston and returns consistent API responses (e.g., `401 Unauthorized` for invalid tokens). 4. **Tests**: Jest unit tests covering 92% coverage for auth flows, including edge cases like concurrent login attempts and token refresh scenarios. 5. **CI/CD Hooks**: Pre-commit hooks for linting (ESLint + Prettier) and a GitHub Actions workflow that auto-deploys to AWS ECS on `main` branch merges. ### Output Artifacts - **Dockerfile**: Multi-stage build with Alpine-based runtime (image size: 45MB). - **Terraform Module**: Auto-generated infrastructure-as-code for AWS ECS with load balancer and RDS Postgres. - **README.md**: Includes setup instructions, API docs (Swagger), and deployment steps. ### Validation Results - Linting: ✅ 0 errors (ESLint strict mode) - Tests: ✅ 18/19 passed (1 flaky test flagged for review) - Security Scan: ✅ No high/critical vulnerabilities (Snyk report attached). Next steps: Run `npm run test:integration` to validate the full auth flow with a live database. The artifact is ready for review in the `zcf-output/` directory.
AI assistant built for thoughtful, nuanced conversation
IronCalc is a spreadsheet engine and ecosystem
ITIL-aligned IT service management platform
Customer feedback management made simple
Enterprise workflow automation and service management platform
Automate your spreadsheet tasks with AI power
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan