Admin Dashboard: Schemas & Autonomous Chat
In ByteBrew V2, each AI workflow is modelled as a Schema — a named configuration that binds an entry agent, sets conversation parameters, and controls access. A schema must be enabled before it can receive chat requests.
Enabling a schema for chat
Section titled “Enabling a schema for chat”To allow a schema to receive messages via the chat API, toggle Chat Enabled in the schema editor:
- Open the Admin Dashboard → Schemas.
- Select the schema you want to activate.
- Toggle Chat Enabled to on.
- Save the schema.
Once enabled, the schema accepts POST /api/v1/schemas/{name}/chat requests. Disabling it returns 403 for all new requests while active sessions drain normally.
Chat API
Section titled “Chat API”# Start or continue a conversation with an enabled schemacurl -N http://localhost:8443/api/v1/schemas/support-handbook/chat \ -H "Content-Type: application/json" \ -d '{"message": "Hello, run the daily report"}'The response streams as Server-Sent Events — the same format as the agent chat API.
Cron & webhook triggers (roadmap)
Section titled “Cron & webhook triggers (roadmap)”Scheduled and event-driven triggers (cron, webhook, on_complete callbacks) are planned for V3. They will let you:
- Run agents on a schedule (
0 9 * * 1-5— weekdays at 9 AM) - Fire an agent from an external webhook (order created, deployment completed)
- Chain trigger results to downstream systems via completion callbacks
Follow the changelog for updates on trigger support.