In active development

Describe a business process. Get a running application.

The Quatrain AI agent turns natural-language requirements into concrete domain models, state machines and screens — then deploys them to your chosen infrastructure. Powered by the chat, skills and OKF bricks, it keeps a persistent second brain of your domain knowledge.

01
Prompt
Natural-language requirements, references, examples.
02
Model
Entities, fields, relations, policies.
03
Flow
State machines and transitions; BPM is on the roadmap.
04
UI
Auto-generated forms and dashboards (CoreUX).
05
Deploy
Any infrastructure, via the Tycho CLI.
Lifecycle aware

Business objects with a memory.

Every entity carries its lifecycle. The agent reuses the same state-machine primitives developers use, so no-code and code converge. A full BPM engine is on the roadmap.

  • · Long-running processes with persisted state
  • · Conversational memory via the chat brick
  • · Second brain: OKF knowledge base accumulated over time
  • · Fine-grained audit trails
  • · Role-based gates and SLAs
agent > transcript
userBuild me an expense approval app with two-level validation and SLA escalation.
agentGenerated Expense, workflow expense.review with states draft → manager → finance → paid, and an SLA escalation hook at 48h.
userDeploy to staging on Kubernetes.
agentDeployed. Dashboard at /admin/expenses.

Want to see the building blocks?

Browse components