> ## Documentation Index
> Fetch the complete documentation index at: https://productlasso.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Enrich products

> Take partial product data and get back enriched records with missing fields filled in, sourced and cited.

Enrich takes partial product records and fills in missing fields using Lasso's product intelligence. Each enriched field comes with a **basis** — citations, reasoning, and confidence — so you know exactly where the data came from.

No tables or files needed. Pass in what you have, get back complete product records.

## Request body

<ParamField body="items" type="object[]" required>
  Array of partial product records to enrich (max 50).

  <Expandable>
    <ParamField body="items[].data" type="object" required>
      Key-value pairs of known product data, e.g. `{ "name": "Sony WF-1000XM5", "sku": "WF1000XM5/B" }`.
    </ParamField>
  </Expandable>
</ParamField>

<ParamField body="schema_id" type="string">
  Reference an existing schema created via `POST /v1/schemas`. Mutually exclusive with `columns`.
</ParamField>

<ParamField body="columns" type="object[]">
  Inline column definitions for the target schema. Mutually exclusive with `schema_id`. If neither is provided, Lasso uses a default product schema.

  <Expandable>
    <ParamField body="columns[].key" type="string" required>Unique column key.</ParamField>
    <ParamField body="columns[].label" type="string" required>Display label.</ParamField>
    <ParamField body="columns[].type" type="string" required>Column type (`text`, `number`, `url`, `tags`, etc.).</ParamField>
  </Expandable>
</ParamField>

<ParamField body="context" type="string">
  Additional context for the AI, e.g. `"These are consumer electronics for a European e-commerce catalog"`.
</ParamField>

<ParamField body="model" type="string" default="auto">
  AI model to use.
</ParamField>

<ParamField body="use_glossary" type="boolean" default="false">
  Apply your company glossary terms during enrichment for consistent terminology.
</ParamField>

<ParamField body="web_search" type="boolean" default="true">
  Whether to use web search for enrichment. When `false`, the AI fills fields from its own knowledge and **no basis/citations are returned** in the response. Useful for fast, low-cost enrichment when sourcing isn't needed.
</ParamField>

<ParamField body="thinking" type="string" default="medium">
  Controls the depth of AI reasoning. Affects both quality and credit cost per item.

  * `hard` — Most capable model, best for complex or ambiguous products. **4 credits/item.**
  * `medium` — Balanced speed and quality. **2 credits/item.** *(default)*
  * `low` — Fastest, suitable for straightforward lookups. **1 credit/item.**
</ParamField>

<ParamField body="webhook_url" type="string">
  If provided, Lasso returns `202` immediately and delivers results via webhook. Also triggered automatically for batches larger than 5 items.
</ParamField>

## Response

<ResponseField name="id" type="string">Enrichment job identifier.</ResponseField>
<ResponseField name="status" type="string">`completed` for sync, `processing` for async.</ResponseField>

<ResponseField name="items" type="object[]">
  Array of enriched product records.

  <Expandable>
    <ResponseField name="items[].data" type="object">Complete product data with all schema columns filled in.</ResponseField>

    <ResponseField name="items[].basis" type="object[]">
      Per-field citations and reasoning for each enriched value. **Only present when `web_search` is `true`.**

      <Expandable>
        <ResponseField name="basis[].field" type="string">The column key this basis applies to.</ResponseField>
        <ResponseField name="basis[].citations" type="object[]">Sources used: `{ url, title, excerpt }`.</ResponseField>
        <ResponseField name="basis[].reasoning" type="string">AI explanation of how it determined the value.</ResponseField>
        <ResponseField name="basis[].confidence" type="string">`high`, `medium`, or `low`.</ResponseField>
      </Expandable>
    </ResponseField>
  </Expandable>
</ResponseField>

<ResponseField name="credits_used" type="number">Credits consumed. Cost per item depends on `thinking`: `hard` = 4, `medium` = 2, `low` = 1.</ResponseField>

<RequestExample>
  ```typescript TypeScript theme={null}
  const result = await client.enrich({
    items: [
      { data: { name: "Sony WF-1000XM5", sku: "WF1000XM5/B" } },
      { data: { name: "AirPods Pro 2", brand: "Apple" } },
    ],
    columns: [
      { key: "name", label: "Product Name", type: "text" },
      { key: "brand", label: "Brand", type: "text" },
      { key: "price", label: "Price", type: "number" },
      { key: "description", label: "Description", type: "text" },
      { key: "features", label: "Features", type: "tags" },
    ],
    context: "Consumer electronics for an e-commerce catalog",
  });

  for (const item of result.items) {
    console.log(item.data.name, "—", item.data.price);
    for (const b of item.basis) {
      console.log(`  ${b.field}: ${b.confidence} (${b.citations.length} sources)`);
    }
  }
  ```

  ```python Python theme={null}
  result = client.enrich(
      items=[
          {"data": {"name": "Sony WF-1000XM5", "sku": "WF1000XM5/B"}},
          {"data": {"name": "AirPods Pro 2", "brand": "Apple"}},
      ],
      columns=[
          {"key": "name", "label": "Product Name", "type": "text"},
          {"key": "brand", "label": "Brand", "type": "text"},
          {"key": "price", "label": "Price", "type": "number"},
          {"key": "description", "label": "Description", "type": "text"},
          {"key": "features", "label": "Features", "type": "tags"},
      ],
      context="Consumer electronics for an e-commerce catalog",
  )

  for item in result["items"]:
      print(item["data"]["name"], "—", item["data"]["price"])
      for b in item["basis"]:
          print(f"  {b['field']}: {b['confidence']} ({len(b['citations'])} sources)")
  ```

  ```bash cURL theme={null}
  curl -X POST "https://hub.banditshq.com/api/v1/enrich" \
    -H "Authorization: Bearer lasso_..." \
    -H "Content-Type: application/json" \
    -d '{
      "items": [
        { "data": { "name": "Sony WF-1000XM5", "sku": "WF1000XM5/B" } }
      ],
      "columns": [
        { "key": "name", "label": "Product Name", "type": "text" },
        { "key": "brand", "label": "Brand", "type": "text" },
        { "key": "price", "label": "Price", "type": "number" },
        { "key": "description", "label": "Description", "type": "text" }
      ]
    }'
  ```
</RequestExample>

<ResponseExample>
  ```json Response theme={null}
  {
    "id": "enrich_abc123",
    "status": "completed",
    "items": [
      {
        "data": {
          "name": "Sony WF-1000XM5",
          "brand": "Sony",
          "price": 179.99,
          "description": "Premium truly wireless earbuds with industry-leading noise cancellation, LDAC codec support, and 30-hour battery life.",
          "features": ["noise cancelling", "LDAC", "30h battery", "IPX4"]
        },
        "basis": [
          {
            "field": "price",
            "citations": [
              {
                "url": "https://www.sony.com/en/headphones/wf-1000xm5",
                "title": "Sony WF-1000XM5",
                "excerpt": "Buy WF-1000XM5 Wireless Noise Cancelling Earbuds — $179.99"
              }
            ],
            "reasoning": "Price confirmed across Sony official store and major retailers.",
            "confidence": "high"
          },
          {
            "field": "features",
            "citations": [
              {
                "url": "https://www.rtings.com/headphones/reviews/sony/wf-1000xm5",
                "title": "Sony WF-1000XM5 Review - RTINGS",
                "excerpt": "Features include active noise cancellation, LDAC codec, 30h battery, IPX4 water resistance."
              }
            ],
            "reasoning": "Feature list compiled from manufacturer specs and third-party reviews.",
            "confidence": "high"
          }
        ]
      }
    ],
    "credits_used": 2
  }
  ```
</ResponseExample>