Prerequisites
- Node.js 22.3 or later.
- An OpenAI API key for model inference.
Create the app
calculator.ts tool, chat page,
AG-UI route, and smoke eval.
Configure inference
Set the API key in the terminal where you run Veryfront:openai/gpt-5.4-nano. Veryfront sends the requests directly
to OpenAI.
Run the app
Verify it worked
Open the local URL printed by the CLI and ask:16.
Run the eval
Stop the dev server withCtrl+C, then run:
Add local delegation
Keep using the same provider key and project. Add a specialist that identifies the important facts:agents/assistant.ts with an orchestrator that can call both
specialists:
delegates list exposes the specialists as agent_researcher and
agent_writer tools. It does not create hosted child runs or require a
Veryfront account.
If the development server is still running, keep using it. If you stopped the
server for the eval, start it again:
agent_researcher and agent_writer, then returns one
combined recommendation. For delegation controls and workflow-based
coordination, see Multi-agent.
Next steps
- Use another inference provider, including Anthropic, Google, Ollama, LM Studio, or a built-in local model.
- Self-host the app in your own environment.
- Use the Veryfront Cloud AI Gateway and deploy with Veryfront Cloud.
- Continue with Create project, or consult the Agent, Tool, and Chat references.