Automations

Automations allow you to create event-driven workflows that automatically execute actions when data in your Shard tables changes. They use user-defined rules to match specific conditions and can trigger multiple actions - like webhooks.

How Automations Work

Automations monitor changes to your Shard tables and execute when:

1. A row is inserted or updated in a specified table
2. The data matches your predicate (filtering condition)
3. The automation is enabled

When these conditions are met, the automation executes all configured actions in sequence.

Key Components

• Source Table: The table to monitor for changes
• Predicate: A JSON Logic expression that filters which rows trigger the automation
• Triggers: Choose whether to run on inserts, updates, or both
• Actions: One or more tasks to execute (webhooks, create rows, etc.)
• Execution History: View logs of every automation run with status and details

Creating an Automation

In the Shard UI

1. Navigate to your Shard and open a table. From the control bar go More -> Automations

2. Click “New” to create a new automation

3. Configure the basics:

   • Name: Give your automation a descriptive name
   • Triggers: Enable “On Insert” and/or “On Update” based on when you want it to run. You can also Enable/Disable the automation from here.

4. Build your conditions using the visual query builder:

   • Add rules to match specific field values
   • Combine rules with AND/OR logic
   • Use operators like equals, contains, greater than, etc.
   • Example: status = "pending" AND priority > 5

5. Add actions that should execute when the predicate matches:

   • Webhook: Send HTTP requests to external endpoints
     • Configure URL, method (GET/POST/PUT/etc.), headers
     • Choose what data to send: just the row, full context, or custom payload
   • More coming soon

6. Enable the automation and click Save

Example Use Cases

• Webhook notification: When a new user signs up (on_insert), send their data to your CRM via webhook
• Status tracking: When an order status changes to “shipped” (on_update), create a tracking record in another table
• Alert system: When an error log with severity = "critical" is created, trigger a webhook to your alerting service
• Data pipeline: Automatically transform and copy data between tables based on conditions

Monitoring Executions

Each automation execution is logged with:

• Status: ok (succeeded), skipped (didn’t match predicate), or failed (error occurred)
• Actions Ran: How many actions were executed
• Execution Log: Detailed output including which actions succeeded/failed and any error messages
• Timestamps: When the execution started and completed

View execution history from the Automations tab to debug issues and monitor performance.

Integrations

Automations can currently integrate with any external service that accepts webhooks. We are working on adding more direct integrations. Have something you want to see? Let us know in Github Discussions.