vibe-coding-yaml

dredd-us's avatarfrom dredd-us

AI outputs YAML descriptors instead of code, CLI executes deterministically for 100% auditable behavior with zero slop. Use when auditability is critical, generated code review needed, or deterministic execution required. YAML serves as intermediate representation. Triggers on "auditable generation", "YAML output", "deterministic execution", "code review", "vibe coding".

⭐0stars🔀0forks📁View on GitHub🕐Updated Nov 20, 2025
[Code Generation]

When & Why to Use This Skill

The Vibe Coding YAML skill revolutionizes AI-assisted programming by shifting from direct code output to structured YAML descriptors. This approach ensures 100% auditable behavior and deterministic execution, effectively eliminating 'AI slop' and hallucinations. By using YAML as an intermediate representation, developers can review logic-heavy descriptors before a CLI generates the final source code, making the development process more transparent, reproducible, and version-control friendly.

Use Cases

  • UI Component Generation: Define UI elements, props, and styles in YAML to ensure consistent, production-ready frontend components without manual refactoring.
  • Infrastructure as Code (IaC): Generate auditable cloud resource descriptors that can be deterministically deployed, ensuring security and compliance in DevOps workflows.
  • API Schema & Endpoint Mapping: Create standardized API definitions in YAML that the CLI can compile into backend routes and validation logic with zero variance.
  • Regulated Software Development: Implement a rigorous review process where auditors verify structured YAML logic instead of parsing complex AI-generated code blocks.
  • Automated Documentation: Use YAML descriptors as a single source of truth to generate both functional code and technical documentation simultaneously.
namevibe-coding-yaml
descriptionAI outputs YAML descriptors instead of code, CLI executes deterministically for 100% auditable behavior with zero slop. Use when auditability is critical, generated code review needed, or deterministic execution required. YAML serves as intermediate representation. Triggers on "auditable generation", "YAML output", "deterministic execution", "code review", "vibe coding".

Vibe Coding YAML

Purpose

AI generates YAML descriptors instead of code, CLI executes deterministically for 100% auditable behavior with zero slop.

When to Use

  • Auditable code generation
  • Deterministic outputs required
  • Generated code review needed
  • UI component generation
  • Infrastructure as code
  • Compliance requirements

Core Instructions

Pattern

AI outputs YAML:

component:
  type: Button
  props:
    text: "Submit"
    variant: "primary"
    onClick: "handleSubmit"
  styles:
    padding: "12px 24px"
    borderRadius: "8px"
    backgroundColor: "#007bff"

CLI executes deterministically:

generate-component component.yaml > Button.tsx

Benefits

  • 100% auditable: Review YAML, not generated code
  • Zero slop: Deterministic execution
  • Versionable: YAML in Git
  • Reproducible: Same YAML = same output
  • Reviewable: Easier to review descriptors than code

Example Workflow

# 1. AI generates descriptor
descriptor = {
    'type': 'api_endpoint',
    'method': 'POST',
    'path': '/users',
    'handler': 'createUser',
    'validation': {
        'email': 'required|email',
        'name': 'required|string'
    }
}

# 2. Save as YAML
with open('endpoint.yaml', 'w') as f:
    yaml.dump(descriptor, f)

# 3. Execute deterministically
$ generate-endpoint endpoint.yaml > routes/users.ts

# 4. Review YAML (not generated code)
$ git diff endpoint.yaml

Performance

  • 100% auditable
  • 0% slop rate
  • Deterministic behavior
  • Easy review process

Version

v1.0.0 (2025-10-23)