// About

AI engineering with evidence.

ARCANA is an OTNEL product built for developers who want to use modern AI models without turning engineering work into an opaque chat transcript. The product is terminal-native and centers model choice, scoped tools, operator control, persistent working context, routing visibility, and evidence that can be inspected after a session completes.

Why ARCANA exists

AI coding tools are increasingly capable of reading repositories, executing commands, editing files, calling services, and coordinating longer workflows. That capability is useful, but it also makes it important to understand what was requested, what was authorized, what actually ran, and what changed. ARCANA is designed around that operational need: make agent-assisted work observable and controllable without requiring developers to abandon the terminal workflows they already use.

Product principles

ARCANA keeps model choice open, supports bring-your-own provider keys, treats permissions and execution boundaries as visible product surfaces, and preserves useful session evidence. Local-first use remains available through the Community edition, while hosted infrastructure and advanced service features are separated into paid tiers. ARCANA documentation, pricing, status, and API descriptions are public so developers and automated clients can evaluate the product before integrating it.

Start with the documentation, review pricing, or read the machine-readable agent guide and OpenAPI specification.