Skip to main content
POST
/
v1
/
monitor
/
chatgpt
Monitor ChatGPT Responses
curl --request POST \
  --url https://api.cloro.dev/v1/monitor/chatgpt \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "prompt": "What do you know about Acme Corp?",
  "country": "US",
  "include": {
    "html": true,
    "markdown": true,
    "rawResponse": true,
    "searchQueries": true
  }
}
'
{
  "success": true,
  "result": {
    "text": "The name \"Acme Corporation\" is used in various contexts, both fictional and real. Here's an overview:\n\n🐾 Acme Corporation in Fiction\nThe Acme Corporation is a fictional company featured prominently in Warner Bros. cartoons, particularly the Road Runner/Wile E. Coyote series.",
    "sources": [
      {
        "position": 1,
        "url": "https://www.rippling.com/blog/ai-recruiting?utm_source=chatgpt.com",
        "label": "12 Best AI Recruiting Tools For HR in 2025",
        "description": "Discover the top AI recruiting tools that can help HR professionals streamline their hiring process...",
        "footnote": false,
        "datePublished": "May 22, 2025"
      },
      {
        "position": 2,
        "url": "https://www.index.dev/blog/ai-recruiting-software-hiring-managers?utm_source=chatgpt.com",
        "label": "7 Best AI Recruiting Software for Hiring Managers in 2025",
        "description": "A comprehensive guide to the best AI recruiting software options available for hiring managers...",
        "footnote": false,
        "datePublished": "April 10, 2025"
      },
      {
        "position": 3,
        "url": "https://www.selectsoftwarereviews.com/buyer-guide/ai-recruiting?utm_source=chatgpt.com",
        "label": "10+ Best AI Recruiting Software for 2025: Expert Reviews + Pricing",
        "description": "Expert reviews and pricing information for the best AI recruiting software solutions...",
        "footnote": true
      }
    ],
    "html": "https://storage.cloro.dev/results/c45a5081-808d-4ed3-9c86-e4baf16c8ab8/page-1.html",
    "markdown": "The name \"Acme Corporation\" is used in various contexts, both fictional and real. Here's an overview:\n\n🐾 Acme Corporation in Fiction\nThe Acme Corporation is a fictional company featured prominently in Warner Bros. cartoons, particularly the Road Runner/Wile E. Coyote series.",
    "rawResponse": [
      {
        "type": "resume_conversation_token",
        "token": "eyJhbGciOiJFUzI1Ni...WaVoqQ",
        "conversation_id": "68efc64b-2d74-8012-b87d-28e59f3b8a49"
      },
      {
        "type": "message_start",
        "message": {
          "id": "msg_01JD6Q7H1ZKWM9GD34V5ZW1TYM",
          "role": "assistant",
          "model": "gpt-5"
        }
      },
      {
        "type": "message_delta",
        "delta": {
          "content": [
            {
              "type": "output_text_delta",
              "text_delta": "The name \"Acme Corporation\" is used in various contexts..."
            }
          ]
        }
      },
      {
        "type": "message_stop",
        "reason": "end_turn"
      }
    ],
    "searchQueries": [
      "What is Acme Corporation?",
      "Acme Corp company overview",
      "Acme Corporation products and services"
    ],
    "model": "gpt-5",
    "shoppingCards": [
      {
        "tags": [
          "stylish casual leather sneaker",
          "heritage retro leather sneaker"
        ],
        "products": [
          {
            "title": "Adidas VL Court 3.0",
            "url": "https://www.adidas.com/us/vl-court-3.0-shoes/ID8797.html?utm_source=chatgpt.com",
            "price": "$57.00",
            "featured_tag": "stylish casual leather sneaker",
            "merchant": "adidas + others",
            "image_urls": [
              "https://images.openai.com/static-rsc-1/example.jpg"
            ],
            "rating": 4.7,
            "num_reviews": 10394,
            "id": "3250714974047560249",
            "offers": [
              {
                "merchant_name": "adidas",
                "product_name": "Adidas Women's VL Court 3.0",
                "url": "https://www.adidas.com/us/vl-court-3.0-shoes/ID8797.html?utm_source=chatgpt.com",
                "price": "$57.00",
                "details": "In stock online and nearby, Delivery between Sat - Mon $4.99",
                "available": true,
                "checkoutable": false,
                "price_details": {
                  "base": "$57.00",
                  "total": "$57.00"
                },
                "tag": {
                  "text": "Best price"
                }
              }
            ],
            "rating_grouped_citation": {
              "title": "Adidas Women's VL Court 3.0",
              "url": "https://www.adidas.com/us/vl-court-3.0-shoes/ID8797.html",
              "supporting_websites": [
                {
                  "title": "Adidas Women's VL Court 3.0",
                  "url": "https://www.macys.com/shop/product/adidas-womens-vl-court-3.0-casual-sneakers-from-finish-line?ID=21107988&pla_country=US&CAGPSPN=pla"
                }
              ]
            }
          }
        ]
      }
    ],
    "entities": [
      {
        "type": "product",
        "name": "adidas Grand Court Lo"
      },
      {
        "type": "product",
        "name": "Reebok Club C 85 Vintage"
      },
      {
        "type": "product",
        "name": "Nike Dunk Low Retro SE"
      }
    ],
    "map": [
      {
        "name": "Blue Bottle Coffee",
        "category": "Coffee shop",
        "rating": 4.5,
        "review_count": 1234,
        "description": "Specialty coffee roaster known for their meticulous pour-over preparations",
        "directions_url": null,
        "website_url": "https://bluebottlecoffee.com",
        "phone_number": "+1-415-555-0123",
        "position": 1,
        "citations": null
      }
    ],
    "citationPills": [
      {
        "url": "https://example.com/article",
        "label": "Example Article Title",
        "description": "A detailed description of the cited content",
        "domain": "example.com",
        "datePublished": "2025-01-01",
        "citationPillId": 0
      }
    ]
  }
}

​
Overview

The ChatGPT endpoint extracts structured data from ChatGPT with advanced features including shopping cards, brand entities, map entries, raw response data, and query fan-out. It’s the most comprehensive monitoring endpoint, ideal for e-commerce tracking, local business monitoring, and deep analysis.
Web search enabledThis endpoint automatically enables ChatGPT’s web search mode for all requests, ensuring responses include current information from the web with proper source citations.

​
Unique features

  • Sources: Sources include publication dates and footnote indicators for distinguishing primary from secondary references
  • Shopping cards: Automatically extracts structured product information with pricing, ratings, offers, and commercial details
  • Entity extraction: Identifies and extracts named entities like products, brands, and concepts mentioned in responses
  • Map entries: Extracts business and place information including ratings, reviews, contact details, and location data
  • Citation pills: Captures inline citations with rich metadata including labels, descriptions, domains, and publication dates
  • Raw response access: Includes the complete streaming response payload for advanced debugging
  • Query fan-out insights: Reveals the actual query fan-out ChatGPT used to generate responses

​
Request parameters

Uses common parameters. ChatGPT-specific options:
  • include.rawResponse (boolean): Include raw streaming response events. Defaults to false (+2 credits)
  • include.searchQueries (boolean): Include search model queries (fan-out queries) that ChatGPT uses internally to gather information. Defaults to false (+2 credits)

​
Sources

Sources are references extracted from ChatGPT’s sources modal. They follow the common sources structure with additional ChatGPT-specific fields.

​
ChatGPT-specific source fields

FieldTypeDescription
footnotebooleanWhether this source appears as a footnote (in a subsequent list in the sources modal)
datePublishedstringThe publication date of the source, if available (e.g., “May 22, 2025”)

​
Example response with sources

{
  "success": true,
  "result": {
    "text": "Here's what I found about AI recruiting tools...",
    "sources": [
      {
        "position": 1,
        "url": "https://www.rippling.com/blog/ai-recruiting?utm_source=chatgpt.com",
        "label": "12 Best AI Recruiting Tools For HR in 2025",
        "description": "Discover the top AI recruiting tools that can help HR professionals streamline their hiring process...",
        "footnote": false,
        "datePublished": "May 22, 2025"
      },
      {
        "position": 2,
        "url": "https://www.index.dev/blog/ai-recruiting-software-hiring-managers?utm_source=chatgpt.com",
        "label": "7 Best AI Recruiting Software for Hiring Managers in 2025",
        "description": "A comprehensive guide to the best AI recruiting software options available for hiring managers...",
        "footnote": false,
        "datePublished": "April 10, 2025"
      },
      {
        "position": 3,
        "url": "https://www.selectsoftwarereviews.com/buyer-guide/ai-recruiting?utm_source=chatgpt.com",
        "label": "10+ Best AI Recruiting Software for 2025: Expert Reviews + Pricing",
        "description": "Expert reviews and pricing information for the best AI recruiting software solutions...",
        "footnote": true
      }
    ]
  }
}
ChatGPT doesn’t provide sources for certain promptsWhen web search is triggered but no sources are available, the sources field will return an empty array.

​
Shopping cards

Shopping cards are automatically extracted when ChatGPT returns product or commercial information. No additional parameters are required - shopping cards are included by default when available in the response.

​
Shopping card structure

Each shopping card contains:
FieldTypeDescription
tagsarrayCategory tags for the shopping card (e.g., [“electronics”, “deals”])
productsarrayArray of product information objects

​
Product information

Each product includes comprehensive e-commerce data:
FieldTypeDescription
titlestringProduct name
urlstringProduct page URL with ChatGPT attribution
pricestringCurrent price (e.g., “$57.00”)
featured_tagstringProduct category or style tag
merchantstringMerchant information (e.g., “adidas + others”)
image_urlsarrayArray of product image URLs (multiple sizes)
ratingfloatProduct rating (0-5 scale)
num_reviewsintegerNumber of reviews
idstringUnique product identifier
offersarrayArray of shopping offers from different merchants
rating_grouped_citationobjectRating source information with supporting websites

​
Offer information

Each offer contains merchant-specific details:
FieldTypeDescription
merchant_namestringMerchant name (e.g., “adidas”, “Macy’s”)
product_namestringProduct name as listed by merchant
urlstringOffer URL with ChatGPT attribution
pricestringOffer price (e.g., “$57.00”)
detailsstringStock and delivery information
availablebooleanOffer availability status
checkoutablebooleanWhether offer can be checked out directly
price_detailsobjectDetailed price breakdown
tagobjectPromotional tag (e.g., “Best price”)

​
Price details

Offers include detailed price breakdown:
FieldTypeDescription
basestringBase product price
totalstringTotal price including any additional costs

​
Rating citation information

Products include rating source attribution:
FieldTypeDescription
titlestringSource title
urlstringSource URL
supporting_websitesarrayArray of supporting website references

​
Example usage

{
  "prompt": "What are the best sneakers under $100?",
  "model": "CHATGPT",
  "country": "US"
}
Response with shopping cards: 
{
  "success": true,
  "result": {
    "text": "If you're shopping for good sneakers under $100, it's definitely doable...",
    "shoppingCards": [
      {
        "tags": [
          "stylish casual leather sneaker",
          "heritage retro leather sneaker",
          "canvas everyday skate-inspired"
        ],
        "products": [
          {
            "title": "Adidas VL Court 3.0",
            "url": "https://www.adidas.com/us/vl-court-3.0-shoes/ID8797.html?utm_source=chatgpt.com",
            "price": "$57.00",
            "featured_tag": "stylish casual leather sneaker",
            "merchant": "adidas + others",
            "image_urls": [
              "https://images.openai.com/static-rsc-1/AY2CiYu1AtGyrY1sX2_dN8Y7naInL0DvIvB1GO-loo2SdZW-LiT6_5gDEgLFJhrM60Vxg2uSbf8tn6WWltaLtpeZkkaGxjwbTm0PlICUvbMwZhhJBABlb-kVVV8m8bYX01_WnkdZrLO2uKEKcYzx3dt1w-E_i-5S4Qarcvp8-hTIM49zkLoHoZcVobqOx8CrulkNdcAoTyegDV532upAMzt1gH7PhFfVVQGyuFsWS_1iaY7_M5qdiKhB01R02Btl"
            ],
            "id": "3250714974047560249",
            "rating": 4.7,
            "num_reviews": 10394,
            "offers": [
              {
                "merchant_name": "adidas",
                "product_name": "Adidas Women's VL Court 3.0",
                "url": "https://www.adidas.com/us/vl-court-3.0-shoes/ID8797.html?utm_source=chatgpt.com",
                "price": "$57.00",
                "details": "In stock online and nearby, Delivery between Sat - Mon $4.99",
                "available": true,
                "checkoutable": false,
                "price_details": {
                  "base": "$57.00",
                  "total": "$57.00"
                },
                "tag": {
                  "text": "Best price"
                }
              },
              {
                "merchant_name": "Macy's",
                "product_name": "Adidas Women's VL Court 3.0",
                "url": "https://www.macys.com/shop/product/adidas-womens-vl-court-3.0-casual-sneakers-from-finish-line?ID=21107988&pla_country=US&CAGPSPN=pla&utm_source=chatgpt.com",
                "price": "$60.00",
                "details": "In stock online and nearby, Free delivery between Fri - Mon",
                "available": true,
                "checkoutable": false,
                "price_details": {
                  "base": "$60.00",
                  "total": "$60.00"
                }
              }
            ],
            "rating_grouped_citation": {
              "title": "Adidas Women's VL Court 3.0",
              "url": "https://www.adidas.com/us/vl-court-3.0-shoes/ID8797.html",
              "supporting_websites": [
                {
                  "title": "Adidas Women's VL Court 3.0",
                  "url": "https://www.macys.com/shop/product/adidas-womens-vl-court-3.0-casual-sneakers-from-finish-line?ID=21107988&pla_country=US&CAGPSPN=pla"
                }
              ]
            }
          }
        ]
      }
    ]
  }
}

​
Entities

The entities object is currently not being returned by OpenAI. The field remains in the schema to support potential future availability.
Entities are structured data objects extracted from ChatGPT responses when specific items, products, or concepts are identified. Entities provide detailed information about individual items mentioned in the response, similar to shopping cards but more general in nature.

​
Entity structure

Each entity contains:
FieldTypeDescription
typestringEntity type identifier (e.g., “product”, “software”)
namestringEntity name or title

​
Example usage

{
  "prompt": "What are the best sneakers under $100?",
  "model": "CHATGPT",
  "country": "US"
}
Response with entities: 
{
  "success": true,
  "result": {
    "text": "Here are some great sneaker options under $100...",
    "entities": [
      {
        "type": "product",
        "name": "adidas Grand Court Lo"
      },
      {
        "type": "product",
        "name": "Reebok Club C 85 Vintage"
      },
      {
        "type": "product",
        "name": "Nike Dunk Low Retro SE"
      }
    ]
  }
}

​
Map

Map entries are automatically extracted when ChatGPT returns business or place information (like local businesses, restaurants, stores, etc.). No additional parameters are required - map entries are included by default when available in the response.

​
Map entry structure

Each map entry contains:
FieldTypeDescription
namestringBusiness name
categorystringPrimary business category
ratingfloatBusiness rating (0-5 scale)
review_countintegerNumber of reviews
descriptionstringBusiness description
directions_urlstringURL to get directions to the business
website_urlstringBusiness website URL
phone_numberstringContact phone number
positionintegerPosition/ranking in results
citationsobjectReference citations (if any)

​
Example usage

{
  "prompt": "What are the best coffee shops in San Francisco?",
  "model": "CHATGPT",
  "country": "US"
}
Response with map entries: 
{
  "success": true,
  "result": {
    "text": "Here are some highly-rated coffee shops in San Francisco...",
    "map": [
      {
        "name": "Blue Bottle Coffee",
        "category": "Coffee shop",
        "rating": 4.5,
        "review_count": 1234,
        "description": "Specialty coffee roaster known for their meticulous pour-over preparations",
        "directions_url": null,
        "website_url": "https://bluebottlecoffee.com",
        "phone_number": "+1-415-555-0123",
        "position": 1,
        "citations": null
      },
      {
        "name": "Philz Coffee",
        "category": "Coffee shop",
        "rating": 4.7,
        "review_count": 2156,
        "description": " neighborhood coffee shop famous for their custom-blended coffee",
        "directions_url": null,
        "website_url": "https://philzcoffee.com",
        "phone_number": "+1-415-555-0456",
        "position": 2,
        "citations": null
      }
    ]
  }
}

​
Citation pills

Citation pills are inline citations that appear within the text of ChatGPT responses when ChatGPT references specific sources. No additional parameters are required - citation pills are included by default when available in the response. The markdown field contains these citations as inline markdown links (e.g., [Label](url)), while the text field contains plain text without the links. The citationPills array provides structured metadata about all inline citations.

​
Citation pill structure

Each citation pill contains:
FieldTypeDescription
urlstringThe URL of the citation source
labelstringLabel/title of the citation
descriptionstringDescription of the citation content
domainstringDomain of the citation source (e.g., “example.com”)
datePublishedstringISO 8601 date string of when the source was published
citationPillIdintegerUnique identifier for the citation pill

​
Example usage

{
  "prompt": "What is the best AI software?",
  "model": "CHATGPT",
  "country": "US"
}
Response with citation pills: 
{
  "success": true,
  "result": {
    "text": "**ChatGPT** — Most versatile for writing, reasoning, and general problem-solving.\n**Claude** — Strong focus on safety and ethical reasoning.",
    "markdown": "**ChatGPT** — Most versatile for writing, reasoning, and general problem-solving. [Top 10 Best AI Apps in 2025](https://www.top10.com/best-lists/best-ai-apps)\n**Claude** — Strong focus on safety and ethical reasoning. [Best AI Platforms Compared](https://www.godofprompt.ai/blog/best-ai)",
    "citationPills": [
      {
        "url": "https://www.top10.com/best-lists/best-ai-apps",
        "label": "Top 10 Best AI Apps in 2025",
        "description": "A comprehensive guide to the best AI applications available today",
        "domain": "top10.com",
        "datePublished": "2025-01-15",
        "citationPillId": 0
      },
      {
        "url": "https://www.godofprompt.ai/blog/best-ai",
        "label": "Best AI Platforms Compared",
        "description": "Expert comparison of leading AI platforms and their capabilities",
        "domain": "godofprompt.ai",
        "datePublished": "2025-02-01",
        "citationPillId": 1
      }
    ]
  }
}

​
Response schema

Includes common response fields plus:
FieldTypeDescription
result.rawResponsearrayArray of ChatGPT’s streamed response events (included when include.rawResponse is true)
result.searchQueriesarrayArray of search model queries (fan-out queries) ChatGPT used internally to gather information (included when include.searchQueries is true)
result.shoppingCardsarrayArray of shopping/product cards extracted from response (when available)
result.entitiesarrayArray of entities extracted from response (when available)
result.maparrayArray of business/place map entries extracted from response (when available)
result.citationPillsarrayArray of inline citation pills extracted from response (when available)
result.sourcesarrayArray of sources referenced in the response. See Sources for ChatGPT-specific fields
result.modelstringThe ChatGPT model used to generate the response

Authorizations

​
Authorization
string
header
required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Body

application/json

Request parameters for monitoring ChatGPT responses

​
prompt
string
required

The prompt to send to ChatGPT

Required string length: 1 - 10000
Example:

"What do you know about Acme Corp?"

​
country
string
default:US

Country/region code for localized response

Example:

"US"

​
include
object

Optional flags for including additional response formats

Example:
{
  "html": true,
  "markdown": true,
  "rawResponse": true,
  "searchQueries": true
}

Response

successful ChatGPT monitoring response

​
success
boolean
required
Example:

true

​
result
object
required

ChatGPT's response data