> ## Documentation Index
> Fetch the complete documentation index at: https://docs.markifact.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Audit Assistant Answer

> Audit what the real ChatGPT or Gemini app answers for a prompt in one country and language, scraped from the assistant rather than the API, for AI visibility, AEO and GEO: returns the markdown answ...

Audit what the real ChatGPT or Gemini app answers for a prompt in one country and language, scraped from the assistant rather than the API, for AI visibility, AEO and GEO: returns the markdown answer, the cited sources, the brands named and a link that reproduces the prompt. Only for measuring AI answers, never as a general assistant to draft content or answer the user.

| | |
| - | - |
| **App** | DataForSEO |
| **Operation ID** | `dataforseo_audit_assistant_answer` |
| **Type** | Action |
| **Connection** | `dataforseo` (required) |
| **Credits per run** | Free |
| **Agent / MCP tool** | Yes |

## Inputs

| Field | Type | Required | Default | Description |
| - | - | - | - | - |
| `platform` | string or SelectableOption | No | `chat_gpt` | Which assistant to query: 'chat\_gpt' (default) or 'gemini'. |
| `prompt` | string | Yes | - | The prompt a user would type into the assistant, e.g. 'best CRM for small businesses' (up to 2000 characters). |
| `location` | string or SelectableOption | No | `2840` | Country or city the assistant is queried from: a two-letter country code (us, gb, de...), a DataForSEO location code (2840 = United States) or a full location name such as 'London,England,United Kingdom'. Defaults to the United States. |
| `language` | string or SelectableOption | No | `en` | Language code of the conversation, e.g. en, de, es, fr, nl. Defaults to en. |
| `force_web_search` | boolean | No | `False` | chat\_gpt only: force ChatGPT to search the web and cite live sources instead of letting it decide. |

### SelectableOption

| Field | Type | Required | Default | Description |
| - | - | - | - | - |
| `value` | string | Yes | - | |
| `label` | string | Yes | - | |

## Output

**Type**: `List[Dict]`

Returns a single row: platform, prompt, model, answer (markdown), sources (list of title, domain, url, source\_name, snippet, publication\_date), cited\_domains (comma-separated unique domains), search\_results (list of title, domain, url, description), brand\_entities (list of title, category, urls), fan\_out\_queries, check\_url (link that reproduces the prompt in the assistant), datetime.

**Fields**: `platform`, `prompt`, `answer`, `sources`, `cited_domains`, `brand_entities`, `check_url`

**Example**:

```json theme={"dark"}
[
  {
    "platform": "chat_gpt",
    "prompt": "best backlink checker tools",
    "answer": "My top picks are Ahrefs, Semrush and Moz...",
    "cited_domains": "ahrefs.com, semrush.com, moz.com"
  }
]
```


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.