AI by Zapier prompts and structured output

Home Tools Platforms Zapier AI by Zapier prompts and structured output

Use an AI step to transform text into declared fields before routing it onward.

Workflow & automation · Changing · Documentation checked 2026-10-04

Zaps, AI by Zapier, Tables, and connected app actions; standalone Agents is migrating.

Paid plan and model tier affect AI features. Tools and knowledge require supported Advanced / Premium models.

The worked example is illustrative and has not been tested in a signed-in product account.

What it does and how it connects

Use an AI step to transform text into declared fields before routing it onward.

AI by Zapier can transform input into declared output fields. Current documentation is changing as standalone Agents moves into AI by Zapier. The prompt-only step should be distinguished from a tool-capable step; plan and model-tier access need confirmation in the current account.

Walkthrough

Use the product named in the access panel. The steps below combine documented controls with the editorial example.

  1. Add an AI by Zapier step to a sample Zap.
  2. Select an available model / provider path and supply controlled input fields.
  3. Define the required output fields and write a focused prompt.
  4. Use Preview / test, inspect the output, and validate it before any downstream action.

Worked example

Classify a fictional workshop question into topic and needs_review. Include a question absent from the approved agenda. The output should preserve uncertainty rather than invent a date to fill a declared field.

Classify the provided workshop question using the approved agenda text. Return topic, source_reference, missing_details, and needs_review. If no supported answer exists, mark needs_review and leave unsupported details empty. Do not infer registration, consent, or identity. No tools or external actions are needed.

Inspect and verify the result

Check schema shape, allowed values, and factual support independently. Compare the output with the original text. Test ambiguous and absent answers as well as a known answer.

Limits and troubleshooting

Official articles differ in how they describe Free / paid access during this transition; confirm available tiers in your account. Model-generated output does not become reliable merely because it fits fields. Do not assume the step learns from earlier runs.

Data, permissions, and actions

Input and prompt context go to the configured model provider under the relevant terms. Bring-your-own-key billing is separate from platform task usage. Keep keys in supported connections rather than prompt fields.

Tools, knowledge, browsing, and Agents migration

Understand when AI by Zapier chooses tools rather than simply transforming an input.

Explore Tools, knowledge, browsing, and Agents migration

A fixed question-to-draft Zap

Turn a synthetic question into declared fields and a private response draft.

Explore A fixed question-to-draft Zap

A tool agent and a reviewed action

Use a supported AI by Zapier tool configuration with an actual pause before a sensitive action.

Explore A tool agent and a reviewed action

Permissions and meaningful approvals

Decide what a system may read, what it may change, and when someone needs to authorize a concrete action.

Explore Permissions and meaningful approvals

Sources and verification

Documentation checked 2026-10-04. Examples are illustrative unless described otherwise.

AI by Zapier: prompts and output fields · Zapier

Share this page