vibeflow

# Vibeflow

[![npm version](https://img.shields.io/npm/v/setup-vibeflow)](https://www.npmjs.com/package/setup-vibeflow)
[![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](LICENSE)
[![Website](https://img.shields.io/badge/vibeflow.run-docs%20%26%20examples-8A2BE2)](https://vibeflow.run)

### Stop vibe-coding blind. Spec it, then ship it.

AI agents write code fast — but without specs, they write the **wrong** code fast. Vibeflow adds a thinking layer before coding: you define **what** to build with specs, guardrails, and quality gates. The agent implements following your project's real patterns.

> Works with **Claude Code** (plugin), **Cursor** (rules + skills), and **GitHub Copilot** (prompts + agents).

## 3 commands to start

```
analyze              → scans your codebase, builds .vibeflow/ knowledge base
gen-spec "feature"   → generates spec with Definition of Done, scope, patterns
implement <spec>     → implements with guardrails (budget, DoD, tests)
```

Run `analyze` once, then `gen-spec` → `implement` for each feature. That's it.

## The pipeline

> **Think** before you code. **Build** from specs. **Verify** against DoD.

`analyze` → `discover` → `gen-spec` → `implement` → `audit`

| Command | What it does | When to use |
|---------|-------------|-------------|
| **analyze** | Deep-scans codebase → `.vibeflow/` knowledge base | Setup, or after major changes |
| **discover** | Dialogue that turns a vague idea into a PRD | Idea isn't clear yet |
| **gen-spec** | Generates spec with binary DoD, scope, anti-scope | Ready to specify |
| **implement** | Implements from spec with guardrails (budget, DoD, patterns) | Spec approved, ready to code |
| **prompt-pack** | Self-contained prompt for another agent/session | Need to delegate |
| **audit** | Verifies DoD + pattern compliance + tests | After implementation |

**Utility:** `quick` (fast-track small tasks), `teach` (update knowledge base), `stats` (audit trends).

## Editions

Each edition adap