ChatGPT

ChatGPT API

Send a prompt to ChatGPT and get the answer it gives: the full Markdown answer, every page it cites with title, publisher and date, the web searches it ran, and the products it recommended. Built for AI visibility and GEO monitoring: track which brands and pages ChatGPT recommends for the prompts your buyers ask. Costs 2 credits per answered prompt; failed prompts are free.

POST/api/v1/chatgpt/prompt

Authorizations

Authorizationstringheaderrequired

Bearer authentication header of the form Bearer <token>, where <token> is your Scavio API key (e.g. Bearer sk_live_your_key).

Body

application/json

promptstringrequired

The prompt, up to 1024 characters. Ask it the way your buyers would.

Example: What are the best running shoes for beginners?

modelstring

Model to answer with. Defaults to auto, the model ChatGPT picks by default.

Example: auto

countrystring

2-letter country code the answer is localized for. Defaults to us.

Example: us

include_htmlboolean

Also return the answer rendered as HTML in answer.html. Defaults to false.

Example: false

Request

from scavio import ScavioClient

client = ScavioClient(api_key="sk_live_your_key")

res = client.chatgpt.prompt("What are the best running shoes for beginners?")
data = res["data"]

print(data["answer"]["markdown"][:500])
for s in data["sources"]:
    print(s["source"], s["url"])
print([p["title"] for p in data["shopping"]])

Response

"platform": "chatgpt",

"prompt": "What are the best running shoes for beginners?",

"model": "gpt-6",

"response_time": 28081

}

from scavio import ScavioClient client = ScavioClient(api_key="sk_live_your_key") res = client.chatgpt.prompt("What are the best running shoes for beginners?") data = res["data"] print(data["answer"]["markdown"][:500]) for s in data["sources"]: print(s["source"], s["url"]) print([p["title"] for p in data["shopping"]])