Managing Workflows

Workflows can be activated or deactivated at any time through conversation. Tell your AI employee to "pause the hourly lead enricher" or "activate the weekly LinkedIn post" and it will toggle the workflow state by name. Deactivated workflows do not run but retain their configuration — they can be reactivated later without losing any settings. This is useful for temporary pauses or seasonal automations that only run during certain periods.

Before deploying a workflow in production, you can test it manually. Ask the AI to run a test execution, which triggers the workflow once and shows you the results. This lets you verify that each step works correctly before enabling the automated schedule. Testing is especially important for workflows that send emails, process payments, or interact with external systems.

Every workflow execution is logged with timestamps, status (success or failure), and execution details. Ask your AI employee to show the execution history for any workflow to see when it ran, how long it took, and whether it completed successfully. This is essential for troubleshooting workflows that are not behaving as expected and for verifying that scheduled automations are running on time.

Workflows can be modified at any time. Tell the AI what you want to change — "change the hourly lead enricher to run every 2 hours instead" or "add a Slack notification when the daily drafter finishes." The AI patches the workflow in place (no delete-and-recreate), preserving the existing setup. For major changes, the AI may suggest building a new workflow rather than modifying the existing one, to avoid disrupting current operations.