> ## Documentation Index
> Fetch the complete documentation index at: https://incredible-42686482-cursor-create-detailed-ai-agent-cookboo.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# JSON Extraction Tool

> Use a tool to extract structured JSON from text

Illustrates schema-driven extraction and validation.

### 1) Provide extraction tools and ask for structured data

```bash
curl -X POST "https://api.incredible.one/v1/chat-completion" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $INCREDIBLE_API_KEY" \
  -d '{
    "model": "small-1",
    "stream": false,
    "messages": [{"role": "user", "content": "Extract structured company M&A info from: Apple acquired Beats for $3B."}],
    "functions": [
      {
        "name": "extract_business_info",
        "description": "Extract and structure business information from unstructured text",
        "parameters": {
          "type": "object",
          "properties": {
            "company": {"type": "string"},
            "industry": {"type": "string"},
            "employees": {"type": "string"},
            "location": {"type": "string"},
            "revenue": {"type": "string"},
            "founded": {"type": "string"}
          },
          "required": ["company", "industry", "location"]
        }
      }
    ]
  }'
```

The response includes a `type: "function_call"` with `function_call_id` and extracted fields. Execute your extraction locally if needed, then send the result.

### 2) Send back the extracted JSON

```bash
curl -X POST "https://api.incredible.one/v1/chat-completion" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $INCREDIBLE_API_KEY" \
  -d '{
    "model": "small-1",
    "stream": false,
    "messages": [
      {"role": "user", "content": "Extract structured company M&A info from: Apple acquired Beats for $3B."},
      {"type": "function_call", "function_call_id": "CALL_ID_FROM_STEP_1", "function_calls": [{
        "name": "extract_business_info",
        "input": {"company": "Apple", "industry": "Technology", "location": "Cupertino", "revenue": "$3B"}
      }]},
      {"type": "function_call_result", "function_call_id": "CALL_ID_FROM_STEP_1", "function_call_results": [{"company": "Apple", "industry": "Technology", "location": "Cupertino", "financial": {"revenue": "$3B"}}]}
    ],
    "functions": [
      {"name": "extract_business_info", "description": "Extract and structure business information from unstructured text", "parameters": {"type": "object", "properties": {"company": {"type": "string"}, "industry": {"type": "string"}, "employees": {"type": "string"}, "location": {"type": "string"}, "revenue": {"type": "string"}, "founded": {"type": "string"}}, "required": ["company", "industry", "location"]}}
    ]
  }'
```

<Note>
  View source on GitHub → [3\_json\_extraction.py](https://github.com/Norditech-AB/Incredible-API-Cookbook/blob/main/02-function-calling/3_json_extraction.py)
</Note>