Pulse API can transform unstructured document content (like Markdown) into structured JSON output by providing a schema definition in your API request. This allows for precise data extraction tailored to your needs.
Schema Parameter (schema)
The schema parameter uses a simplified format where you directly specify the expected data types for each key. It supports basic types, arrays, and enums without requiring the full OpenAPI 3.0 specification.
Example 1: Transaction Data
{
"transactions": [
{
"date": "date",
"description": "string",
"amount": "float",
"balance": "float",
"accrued_bank_charges": "float|null"
}
]
}
Example 2: Invoice Data
{
"name": "string",
"customer_email": "string",
"street_address_line_1": "string",
"street_address_line_2": "string",
"city": "string",
"state": "string",
"zip_code": "string",
"country": "string",
"total_price.unit": "USD",
"revenue_schedule": [
{
"service_start_date": "date",
"service_term": "integer",
"service_end_date": "date",
"item_name": "string",
"item_description": "string|null",
"integration_item": "string",
"billing_type": "enum(FLAT_PRICE, PER_UNIT)",
"total_price": "float",
"accountingCurrency": "USD",
"quantity": "integer",
"start_date": "date",
"period": "integer",
"#_of_periods": "integer",
"frequency_unit": "enum(YEAR, MONTH, DAY)",
"end_date": "date",
"net_terms": "integer",
"arrears": "boolean|null"
}
]
}
(Note: Types like string, float, integer, date, boolean, null, and simple enum(VALUE1, VALUE2) definitions are used directly)
Experimental Schema Parameter (experimental_schema)
The experimental_schema parameter utilizes a subset of the OpenAPI 3.0 specification for more complex schema definitions.
Supported operations:
type: (Required) Defines the data type (string, number, integer, boolean, array, object).properties: (Required for object type) Defines key-value pairs within an object, following OpenAPI 3.0 formatting.required: (Optional for object type) An array listing mandatory property keys.items: (Required for array type) Defines the schema for elements within an array, following OpenAPI 3.0 formatting.nullable: (Optional) Boolean (true/false, default false). Allows the value to be null.description: (Optional) A string providing context or use-case for the value.
Example: Condensed Profile Schema
{
"type": "object",
"properties": {
"basics": {
"type": "object",
"description": "Basic personal information",
"properties": {
"rawName": { "type": "string", "description": "Full name", "nullable": true },
"email": { "type": "string", "description": "Email address", "nullable": true },
"phone": { "type": "string", "description": "Phone number", "nullable": true },
"location": { "type": "string", "description": "Formatted location text", "nullable": true },
"gender": {
"type": "string",
"description": "Gender identifier",
"enum": ["Male", "Female", "Non-Binary", "Other"],
"nullable": true
}
},
"required": ["rawName", "email"]
},
"experience": {
"type": "array",
"description": "List of work experiences",
"items": {
"type": "object",
"properties": {
"company": { "type": "string", "description": "Company name", "nullable": true },
"position": { "type": "string", "description": "Job title", "nullable": true },
"startDate": { "type": "string", "description": "Start date (ISO format)", "format": "date", "nullable": true },
"endDate": { "type": "string", "description": "End date (ISO format) or null", "format": "date", "nullable": true },
"description": { "type": "string", "description": "Job details", "nullable": true }
},
"required": ["company", "position", "startDate"]
}
}
},
"required": ["basics", "experience"]
}
Schema Prompting
For both schema types, you can enhance extraction accuracy by providing natural language instructions within the schema’s description fields or as a separate schema_prompt parameter. These prompts guide the model in mapping document content to your schema, handling logical expressions, complex enum mapping, calculations, and other nuanced extraction requirements.
