modelriver trigger

Send test async requests from the terminal. Get channel IDs and WebSocket details for monitoring responses.

Overview

The trigger command sends an async AI request to ModelRiver and returns the channel details you need to monitor the response: either via WebSocket or webhook.

Usage

Bash

# Basic trigger
modelriver trigger --workflow my-workflow --message "Test message"
 
# With custom payload (--workflow can be separate or inside the JSON)
modelriver trigger --workflow my-workflow --payload '{"messages": [...], "metadata": {"ticket_id":"TCK-1"}}'
 
# Create webhook to receive response
modelriver trigger --workflow my-workflow --message "Test" --webhook-url https://webhook.site/your-id
 
# Print channel details
modelriver trigger --workflow my-workflow --message "Test" --print-channel

Backend pipeline tip

If the workflow uses a backend pipeline, keep modelriver listen --print running, then complete with:

Bash

modelriver callback --channel-id CHANNEL_ID --data '{"summary":"..."}'

Example output

✓ Async request created
 
Channel Details
{
  "channel_id": "abc-123-def",
  "project_id": "xyz-789",
  "websocket_url": "wss://api.modelriver.com/socket",
  "websocket_channel": "ai_response:xyz-789:abc-123-def",
  "status": "pending"
}
 
> Use --webhook-url to automatically receive responses via webhook
> Or use "modelriver websocket" to connect and receive responses
> For backend pipeline workflows, complete with: modelriver callback --channel-id <id> --data '{...}'

Options

Option	Description
`--workflow <name>`	Workflow name (required unless included in `--payload`)
`--message <text>`	Simple text message
`--payload <json>`	Custom JSON payload
`--webhook-url <url>`	Webhook URL for responses
`--print-channel`	Print channel details
`--verbose`	Show detailed logs

Next steps

WebSocket: Connect and receive the response in real-time
Listen: Monitor all webhook events
Callback: Complete backend pipeline flows
API: Full async request documentation