Get the screen each AI engine showed with its answer

Turn on include.html for ChatGPT, Gemini, Perplexity, Meta AI, Google AI Overviews or AI Mode and get a link to the engine’s own answer screen, saved as a page that opens without an API key.

What you get

• payload.include.html: true on CHATGPT, GEMINI, PERPLEXITY, META_AI, GOOGLE and AIMODE.
• html in the response: status (ok or failed), url and expiresAt. The link works for 30 days after the task finishes.
• The page opens without an API key and loads nothing from outside, so you can link it or embed it in a report.
• The answer fields (text, sources and the rest) arrive as before. A page that fails never fails the task.

How it works

On ChatGPT, Gemini, Perplexity and Meta AI, the question is asked in a real browser on the engine’s own site, and the screen is saved when the answer finishes: the sidebar, the question, the answer with its source chips, and the message box. The answer fields come from that same ask, and a screen that does not match the answer is not saved.

Scripts are removed before the page is saved, and its images, styles and fonts are stored inside the file. The page opens the same way in any browser, without loading anything from the engine.

On Google AI Overviews and AI Mode, the page is the results page Google returned for the task, with its scripts removed. The AI Overview and its source list are saved open, as if Show more and Show all had been pressed, so the whole answer reads without scripts.

When the screen cannot be captured or saved, html is { "status": "failed", "error": "EVIDENCE_FAILED" } and the task completes with its answer. A ChatGPT task with location, CN or HK, or continueFrom, and a Gemini task with location, CN, DE, GB or IR always come back without a page.

Where it helps

• Put the engine’s own screen next to each number in a client report, so readers see the answer the way a user saw it.
• Open the page when a parsed result looks off, and see where the brand or source sat on the screen.
• Keep a dated record of what each engine answered. Pages stay for 30 days, so download the ones you need longer.

Send a request

Send the request with your API key in the Authorization header.

curl -X POST https://api.querying.ai/v1/async/task \
  -H "Authorization: Bearer $QUERYING_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "taskType": "PERPLEXITY",
    "payload": {
      "prompt": "What are the best CRM tools for a 10-person startup?",
      "country": "US",
      "include": { "html": true }
    }
  }'

Request fields

taskTypestringRequired

The engine to run, such as CHATGPT or GOOGLE.

payload.promptstringRequired

The question, written the way a user types it into the chat. Up to 100,000 characters.

payload.countrystring

Two-letter country code of the market to ask from, such as US, KR or JP. Default US.

payload.include.htmlboolean

true adds html, a link to the screen the engine showed with the answer. On CHATGPT, GEMINI, PERPLEXITY, META_AI, GOOGLE and AIMODE. Off by default.

webhook.urlstring

An HTTPS address on your server. The finished task is posted there the moment it completes.

Response

A finished task looks like this, trimmed for length.

{
  "success": true,
  "task": {
    "id": "7be8662e-5f62-48bf-a54a-7b58a2701b7c",
    "taskType": "PERPLEXITY",
    "status": "COMPLETED",
    "createdAt": "2026-10-21T01:20:07.445Z"
  },
  "credits": { "creditsToCharge": 1.25, "creditsCharged": 1.25 },
  "response": {
    "text": "For a 10-person startup, the strongest shortlist is **HubSpot** for a free, easy starting point; **Pipedrive** for a straightforward sales pipeline; **Zoho CRM** for budget-friendly customization; and **Attio** for a flexible, data-driven CRM.[1][2][3]\n\n…",
    "sources": [
      {
        "position": 1,
        "url": "https://slack.com/blog/crm/best-crms-for-startups",
        "label": "Best CRMs for Startups: Tools, Benefits, and Tips"
      },
      {
        "position": 2,
        "url": "https://attio.com/f/best-crm-for-startups",
        "label": "Best CRMs for startups in 2026"
      }
    ],
    "html": {
      "status": "ok",
      "url": "https://api.querying.ai/v1/evidence/1792545851-e8cc8a32dd7f456eb807a0e09e9bc0fa.html",
      "expiresAt": "2026-11-20T01:24:11.000Z"
    }
  }
}

task.statusstring

QUEUED, PROCESSING, then COMPLETED with the answer or FAILED with error.

credits.creditsChargedinteger

Credits charged for the finished task. A failed task charges 0.

response.textstring

The answer as plain text.

response.sourcesarray

The pages the answer cites, in the engine’s order, each with position, url and label.

response.htmlobject

The engine’s screen: status (ok or failed), url (opens without an API key) and expiresAt, 30 days after the task finished. A failed page has error EVIDENCE_FAILED, and the task still completes with its answer.

Receive the result

The request returns the task right away with status QUEUED. Poll GET /v1/async/task/{id} until status is COMPLETED or FAILED, or add webhook.url and the finished task arrives at your server with the same task, credits and response. Results stay available for 24 hours after a task finishes.

curl https://api.querying.ai/v1/async/task/$TASK_ID \
  -H "Authorization: Bearer $QUERYING_API_KEY"

Errors

EVIDENCE_UNSUPPORTEDHTTP 422

include.html is on for an engine that has no page. Only CHATGPT, GEMINI, PERPLEXITY, META_AI, GOOGLE and AIMODE return one.

VALIDATION_ERRORHTTP 400

A field is missing or out of range. details names the field.

MISSING_API_KEYHTTP 401

The Authorization header is missing, or its key is unknown, expired or revoked.

INSUFFICIENT_CREDITSHTTP 402

The balance is below the price of the task. Add credits or upgrade the plan, then send again.

Pricing

On ChatGPT, Perplexity and Meta AI, the page adds 0.25 credits to the engine’s price. The 0.25 is held when you submit and charged only when html.status is ok. On Gemini, Google AI Overviews and AI Mode, the page is free. With the page, a task on any of the six engines costs 1.25 credits, like the Perplexity task in the example.