Ship intent, not code.
A practical paradigm for the age of vibe coding. Document your intent precisely — Why, What, Not — and delegate everything else to AI.
Intent Engineering is the discipline of capturing what to build and what not to build in a single markdown file (INTENT.md), then letting AI handle implementation, verification, and deployment.
It's not a framework. It's not a tool. It's a file and a discipline.
# INTENT — [Project Name]
> status: seed | exploring | clarified | killed
## Why ← Why does this exist?
## What ← What are we building?
## Not ← What will we never do?
## Learnings ← What did we discover along the way?Why defines the reason this project exists. In early stages it's a hypothesis; once validated, it's conviction.
What describes features and user flows concretely, without prescribing implementation.
Not sets hard boundaries AI must not cross — security, scope, quality bars.
Learnings is a living log of experiments and their outcomes, tracking how intent evolves through exploration.
Intent evolves through four states:
seed → exploring → clarified → build
│ │ │
└────────┴────────────┴──→ killed
- seed: Just an idea. Write a hypothesis, start experimenting.
- exploring: Actively validating through prototypes, interviews, research.
- clarified: All sections filled with conviction. Ready to build.
- killed: Evidence says stop. Record why — that knowledge is valuable.
- Create
INTENT.mdat your project root - Pick your starting state (seed / exploring / clarified)
- Fill in Why. Be honest about What and Not — mark unknowns with
(?) - Explore, learn, update. Or kill when evidence says so.
- Once clarified, hand off to AI.
Full guide: intentengineering.dev (or see docs/)
MIT