Aura for developers

Deploy an optimization model, then run it and talk to Aura about the results, from your own backend, or from the terminal.

Quickstart API reference

npm install -g @strangeworks-inc/strangeworks-aura-cli

The CLI administers Aura. Deploy a model, manage projects and conversations, and chat with a project to check it behaves the way you meant before you build against it.

aura deploy
aura project create acme/vrp --name "vrp demo"
aura chat

Get started with the CLI →

npm install @strangeworks-inc/strangeworks-aura-sdk

The TypeScript SDK embeds Aura in a Node backend. Same API, called from your own service, so your users get an answer from the product they already use. One runtime dependency.

const project = await aura.createProject({ name: "vrp demo", fromDeployment: "acme/vrp" });
const reply = await aura.chat(project.id, "Solve this and summarize the result.");

Get started with the TypeScript SDK →

pip install strangeworks-aura-sdk

The Python SDK embeds Aura in a Python backend. Same API, called from your own service, so your users get an answer from the product they already use. Async throughout.

project = await aura.create_project("vrp demo", from_deployment="acme/vrp")
reply = await aura.chat(project.id, "Solve this and summarize the result.")

Get started with the Python SDK →

Next steps

QuickstartFrom an API key to a solved run.

Core conceptsDeployments, projects, conversations, snapshots — what each one is.

Deploy a modelTurn a solver folder into a deployment.

Instructions and skillsTeach the agent your vocabulary, rules, and procedures.

API referenceEvery endpoint on the public /api/v1 surface.