AI Agent Chat

Features

• Actor-per-agent pattern with a coordinating manager Rivet Actor
• Queue-based intake using c.queue.next inside the run loop
• Streaming AI responses sent to the UI as they arrive
• Persistent history stored in Rivet Actor state
• Live status updates via events and polling

Prerequisites

• OpenAI API key set as OPENAI_API_KEY

Implementation

The AgentManager creates and tracks agent actors, while each AI agent Rivet Actor consumes queue messages in run and streams responses with the Vercel AI SDK.

• Actor definitions and queues (src/actors.ts)
• Frontend orchestration (frontend/App.tsx)
• Server entry point (src/server.ts)

Resources

Read more about queues, run handlers, state, and events.