Workflow node reference
AI Response
Generate an AI response to the message. Exact configuration, outputs, execution behavior, errors, limits, and examples for the AI Response workflow node.
What this guide covers
- Category: AI.
- Use it to generate bounded natural-language output from instructions and workflow context.
- 4 configuration fields and 1 documented outputs.
- Output can be incorrect, variable, or unsafe without constraints.
What this node does
Generate an AI response to the message
Use it to generate bounded natural-language output from instructions and workflow context.
When not to use it
Do not use it for deterministic calculations, exact lookups, or high-impact decisions without review.
Prerequisites
- Clear instructions
- Only necessary input context
- A downstream validation or human-review strategy
- AI usage entitlement/Balance
Clear instructions. Only necessary input context. A downstream validation or human-review strategy. AI usage entitlement/Balance
Inputs and configuration fields
| Field | Key | Control / type | Requirement | Default | Accepted values and behavior |
|---|---|---|---|---|---|
Instructions | system_prompt | textarea | Required when shown | No default | Tell the AI how to behave and respond. You are a helpful customer support agent for our company... |
Message to Respond To | input | variable-input | Required when shown | {{context.message.content}} | The message the AI should respond to. Value accepted by the displayed control. Variables supported |
Include Contact Info | include_contact_info | toggle | Optional | true | Give AI access to contact name and history. Value accepted by the displayed control |
support contact | support_contact | text | Optional | No default | In case AI doesn't have specific information about something, redirect customer.. support@myshop.com or call 1-800-SHOES |
Configure the node
- Drag AI Response from AI in the node palette.
- Select the node to open its configuration drawer.
- Complete every required field and any conditional field revealed by your choices.
- Insert variables only from the current trigger or an earlier node on the same path.
- Connect every required output handle to the intended next step.
- Save and run Workflow Test with representative data before activation.
Outputs
Later nodes can use these values only after this node has run successfully on the same execution path.
| Variable | Type | Meaning |
|---|---|---|
{{nodes.ai_response.response}} | string | The generated response |
Execution behavior
- Sends the configured prompt/context to the selected AI behavior.
- Returns generated response text and model/usage metadata defined by the node.
- Runs for real during Workflow Test and contributes to estimated cost.
AI nodes run for real during Workflow Test and can consume billable AI usage. Do not assume an automatic retry unless the execution record explicitly shows one.
Example use case
Draft a concise support reply using the customer's message and an approved policy summary.
Message Received → Classifier → Condition → AI Response → Send TextCommon errors
| Symptom | Resolution |
|---|---|
Response is off-topic | Tighten instructions and provide bounded context. |
Prompt data is missing | Confirm every referenced variable exists on the path. |
AI step fails | Review node error, entitlement, input size, and retry only after identifying the cause. |
Limitations
Output can be incorrect, variable, or unsafe without constraints.
Never include secrets or unnecessary sensitive data.
Output can be incorrect, variable, or unsafe without constraints. Never include secrets or unnecessary sensitive data.