Build with AI (from scratch)
The easiest way to get started. Open any table, click the AI chat on the right sidebar, and describe what you want:- “When a new order comes in, send a Slack notification with the customer name and amount.”
- “Every Monday morning, email me a summary of tasks that are overdue.”
- “When someone submits the feedback form, classify it and update the category field.”
Build with AI (from trigger)
You can also start from a specific trigger and let AI build the actions:- Add a trigger — choose a trigger type (e.g. “When Record Created”) and configure it.
- Add an action — click +, go to Build with AI, and select Run Script.
- In the configuration panel, click Open Editor.
- Pick from the suggestions to get started quickly, or describe what you need in your own words.
- AI generates the JavaScript code. Review it, click Apply.
- Click Test to run the script with real data, then Enable the workflow.
Build manually
You can also build automations step by step using the built-in actions:- Add a trigger — choose a trigger type, select the table, configure fields. Click Test.
- Add an action — choose an action type (Create Record, Send Email, HTTP Request, etc.), map field values using + to insert dynamic variables. Click Generate Preview or Run as Configured.
- Enable — click the toggle in the top-left corner.
Editing a live workflow
Changes are saved as a draft. The live workflow keeps running on the previous version until you click Apply Update.Secrets
Use Secrets to store API keys, credentials, and configuration that an automation needs at runtime. Open the workflow on the editing tab, then click Secrets next to the test controls. Secrets are scoped to the current automation. Script actions can read them from environment variables such asprocess.env.EXTERNAL_API_KEY. Saved values are write-only; enter a new value to update a secret.
Run history
Open a workflow → Run history. The history panel lists each run with its status, time spent, and step details. Click a run to inspect per-step input/output. Use the history filters to narrow the list by status, duration, or date range. For failed runs, use the run detail panel to diagnose or rerun:- Diagnose opens AI Chat with the failed run as context, so you can ask Cuppy to explain the failure.
- Full rerun runs every step again and may create duplicate records, emails, or notifications.
- Resume from failed step is available when Teable can safely continue from the failed step.
Limits
| Item | Limit |
|---|---|
| Email sending rate | 5 / second / base |
| Webhook receiving rate | 50 / second / base, 2 / second / workflow |
| Webhook payload size | 4 MB |
| Run quota & history retention | Varies by plan |
FAQ
Why did I receive an automation run usage alert?
Why did I receive an automation run usage alert?
In Teable Cloud, Teable sends automation run usage notifications to Space Owners when a space is close to or over its run quota, or when runs increase sharply in a short time.For quota reminders, Teable sends a notification when the current billing period reaches 80% or 90% of the available run quota.Pro and above plans may receive alerts after automation runs go over the quota. Teable may remind you about every 6 hours.Teable Cloud may also send critical notifications for unusual usage, such as a large daily increase or a sudden burst in a short time. The system uses recent daily and hourly run patterns, and the exact thresholds may change.The notification opens Space Settings → Billing. When available, it shows which workflow and Base contributed the most runs. Quota reminders are based on runs counted toward your plan limit; unusual usage checks may also consider recent runs with success, failed, or pending status.If the usage is expected, consider buying extra automation quota or changing your plan. If it is unexpected, review the top workflow and adjust or pause the related automation.