Inspect ChatGPT search content and sources

Provide a query and supported language and location codes. Your Agent retrieves structured ChatGPT search content through DataForSEO, separating available answer content, referenced sources and other retrieved webpages.

Your AI

Use with your AI

For tasks with several steps, continue in the AI you already use.

CodexClaude CodeCursorWindsurfCline

View instructions

Use chatgpt-llm-scraper-live-dataforseo to inspect one structured ChatGPT search result through DataForSEO. Reuse the current Toolmu connection and agreed spending scope; when connection setup is needed, follow https://toolmu.com/SKILL.md. Ask only for missing required information. Submit request with all three properties: keyword, language_code and location_code. keyword and language_code must be nonempty strings of at most 256 characters; location_code must be a supported supplier integer code from 1 to 100000000. Use supported codes, not language or location names, and do not treat en/2840 as defaults. When literal plus or percent signs are important, respect the supplier’s documented %2B/%25 query encoding and count the submitted string against the 256-character cap. Add no model, force-search, history or device settings. On success, inspect the whole returned query result and present available answer content. Clearly separate sources actually cited or relied on from retrieved web entries, which can repeat or remain unused. Keep missing references and unknown values explicit; do not infer that a brand or source never appears from their absence in one result. Report model, positions, counts and UTC receipt time only if returned. Positions are result-display context, not quality or visibility scores; receipt time does not date every source or fact. Check important claims against available source pages. Do not promise current facts, complete coverage, a consumer-account replay or ongoing citation/brand tracking.

Task:
For [your search phrase], show the answer people in [country or region] see in [language]. Separate pages the answer used from other pages found during the search but not used.

Paste into your connected AI's conversation.

Connection guide

Up to 0.11 credits / call

The final charge is based on confirmed usage. Unused reserved credits return to your available balance.

DataForSEO

Inspect ChatGPT search content and sources

You receive

A successful request returns one structured query result, with answer content, sources and web entries when provided. Your Agent can distinguish sources cited or relied on in the answer from webpages retrieved but not necessarily used, helping you inspect the answer’s source context.

Start with your task

For [your search phrase], show the answer people in [country or region] see in [language]. Separate pages the answer used from other pages found during the search but not used.

What to know

Search use and citations are not guaranteed. Retrieved webpages can repeat or remain unused. Check important facts and source dates; one query result does not measure ongoing brand visibility or provide complete web coverage.

Inputs & outputs

Inspect the current inputs, availability, and maximum charge before starting.

Input schema

Input schema
Field / Path	Type	Requirement	Meaning
Search request`request`	`object`	Required	Required object containing one query, a supported language code and a supported location code. All three properties are required. Required fields `language_code, location_code, keyword`
Query`request.keyword`	`string`	Required	Required query or search phrase, from 1 to 256 characters in the submitted string. This Tool handles one query per request. Minimum length `1` Maximum length `256`
Search language code`request.language_code`	`string`	Required	Required supported language code as a string of 1–256 characters. en is an English example, not a default; use a code rather than a language name. Minimum length `1` Maximum length `256`
Search location code`request.location_code`	`integer`	Required	Required supported supplier location code, an integer from 1 to 100000000. 2840 is a United States example, not a default. The numeric range does not make every number a supported location. Minimum `1` Maximum `100000000`

View raw JSON

{
  "type": "object",
  "required": [
    "request"
  ],
  "properties": {
    "request": {
      "type": "object",
      "required": [
        "language_code",
        "location_code",
        "keyword"
      ],
      "properties": {
        "keyword": {
          "type": "string",
          "maxLength": 256,
          "minLength": 1
        },
        "language_code": {
          "type": "string",
          "maxLength": 256,
          "minLength": 1
        },
        "location_code": {
          "type": "integer",
          "maximum": 100000000,
          "minimum": 1
        }
      },
      "additionalProperties": false
    }
  },
  "additionalProperties": false
}

Result schema

Output fields have not been documented yet.

Additional notes

For a literal + or % in the query, DataForSEO requires %2B or %25 respectively; the submitted string must still fit the 256-character input limit. See the official ChatGPT LLM Scraper documentation.

Questions about this tool.

Inputs, results and details to know before running it.

Get connected

Supply one query, a supported language code and a supported location code. The query and language-code strings must each be 1–256 characters. For example, albert einstein with en and 2840 requests English search content for the United States; these values are examples, not defaults.

Inspect ChatGPT search content and sources

Inspect ChatGPT search content and sources

You receive

Start with your task

What to know

Inputs & outputs

Input schema

Result schema

Choose a topic

Pick a language and region

Which pages the answer used

Questions about this tool.

You receive

Start with your task

What to know

Inputs & outputs

Input schema

Result schema

Choose a topic

Pick a language and region

Which pages the answer used

Questions about this tool.

What query and market information do I need?

Are all retrieved webpages cited in the answer?

Will every result include web search and fresh citations?

Is the result always a plain-text answer?

Can I choose a model or replay my own ChatGPT session?