Transforms a_parse_template

Parse Template

Turns a [[field: description]] template string into an OpenAI JSON schema for collect_data.

What does it do?

Fills in a text template (like a pre-written message) with dynamic data from the customer or the flow.

How does it work?

You give it a base text with variables like Hello {{name}} and pass in memory or a data object. The node replaces the variables and returns the final text: Hello Juan.

Why use it?

Perfect for generating standardized messages before sending (e.g. order summaries, appointment confirmations or WhatsApp templates) without spending AI tokens.

Inputs

name	type
`template_body`	string

Outputs

name	type
`template_schema`	object
`template_field_names`	array
`field_count`	number

Branches

Fixed branches

out — Output

Configuration

template_field string required default: "template_body"

output_schema_field string required default: "template_schema"

output_fields_field string required default: "template_field_names"

schema_name string required default: "clinical_note"

infer_types boolean required default: false

inference_model string optional

Default configuration

{
  "template_field": "template_body",
  "output_schema_field": "template_schema",
  "output_fields_field": "template_field_names",
  "schema_name": "clinical_note",
  "infer_types": false
}

Emitted events

template.parsed
Template successfully parsed into JSON schema.
template.types_inferred
Field types inferred via LLM.