# Google AI Mode Scraper - AI Answers with Sources (`scrapewise/google-ai-mode-scraper`) Actor Scrape Google AI Mode answers to search queries, no login: full answer in markdown and plain text, cited sources with URL, title and domain, citations per passage, optional brand check, by country and language. US$ 0.035 per answer, errors free. - **URL**: https://apify.com/scrapewise/google-ai-mode-scraper.md - **Developed by:** [Scrapewise Data](https://apify.com/scrapewise) (community) - **Categories:** AI, SEO tools - **Stats:** 2 total users, 1 monthly users, 100.0% runs succeeded, 0 bookmarks - **User rating**: No ratings yet ## Pricing from $29.75 / 1,000 google ai mode answer delivereds This Actor is paid per event. You are not charged for the Apify platform usage, but only a fixed price for specific events. Since this Actor supports Apify Store discounts, the price gets lower the higher subscription plan you have. Learn more: https://docs.apify.com/actors/running/actors-in-store.md#pay-per-event ## What's an Apify Actor? An Actor is a serverless cloud program that runs on the Apify platform. It has two run modes. In Batch mode, an Actor accepts a well-defined JSON input, performs an action which can take anything from a few seconds to a few hours, and optionally produces a well-defined JSON output, datasets with results, or files in key-value store. In Standby mode, an Actor provides a web server which can be used as a website, API, or an MCP server. Apify vocabulary and the platform model are defined once, in the agent quickstart at https://apify.com/agents.md. ## How to integrate an Actor? If asked about integration, you help developers integrate Actors into their projects. You adapt to their stack and deliver integrations that are safe, well-documented, and production-ready. Do not guess an integration path. Every one of them is in the agent quickstart at https://apify.com/agents.md: the Apify MCP server, Agent Skills with the Apify CLI, the JavaScript and Python clients, the REST API, and the account-free path for an agent with no human to sign in. It also carries the rule on stating cost before the first paid run. For examples already wired to this Actor's own input schema, see the [API](#api) section below. Each client library has reference documentation the quickstart does not restate: [JavaScript/TypeScript](https://docs.apify.com/api/client/js/docs.md) (`npm install apify-client`) and [Python](https://docs.apify.com/api/client/python/docs.md) (`pip install apify-client`). # README ## Google AI Mode Scraper Google AI Mode Scraper collects **what Google AI Mode answers to a search query**, without login. Give it a list of queries (best AI coding assistants for developers, is the iPhone 17 worth upgrading from the iPhone 15, Notion vs Obsidian) and get back, for each one, the **full AI Mode answer in Markdown and plain text**, every **source Google cited** (URL, title, domain, site name, snippet, position), and the **citations passage by passage** (which sources back which part of the answer). One clean row per query, ready for a spreadsheet, a dashboard or an AI pipeline. **US$ 0.035 per answer (US$ 35 per 1,000), plus US$ 0.005 once per run. No monthly fee. Error rows are free.** ### At a glance - **Price per answer:** US$ 0.035 on the Free plan, down to US$ 0.02975 on Gold and above, plus US$ 0.005 once per run - **Engine:** Google AI Mode, the "AI Mode" tab of Google Search, logged-out (what a visitor sees) - **Answer:** Markdown with headings, bullet lists, bold and tables, plus a plain text copy - **Sources:** Every cited URL with position, title, domain, site name and snippet, flagged `citedInAnswer`, tracking parameters removed - **Citations:** Each cited passage with its text and the sources behind it - **Location:** Any country with a 2-letter code (US, GB, DE, FR, JP...) and any language; Google localizes brands, stores and sources - **Brand check (optional):** Mentioned or not, mention count, rank, domain cited as a source, same price - **Error rows:** Free, with a `status` and an `errorCode` ### Why this Actor - **Price.** An AI Mode answer with its sources and passage citations costs US$ 0.035 here. Ten thousand answers cost US$ 350 on the Free plan and less on paid plans, with no charge for errors. - **Answers you can read.** `answer` keeps the structure Google shows (section headings, bullet lists, bold product names, comparison tables) as clean Markdown. `answerText` is the same answer as plain text for spreadsheets, NLP and embeddings. - **Real source URLs.** Sources come straight from the page data, so you get the publisher URL, not a Google redirect link, plus the title and the snippet Google used. - **Which source backs which sentence.** `citations` maps each passage of the answer to the sources behind it, so you can see exactly what a site is credited for. - **Inputs that just work.** Country and language are free text: `US`, `us`, `Germany`, `en-GB` all work, and an unknown value falls back to US with a log line instead of failing the run. - **Clear failures.** A query that could not be answered becomes a free error row with a reason, so you never pay for an empty result. ### Switching from another Google AI Mode actor? Your existing input works The input names used by other AI Mode and AI search actors are accepted as they are, so you can paste your current input: | You send | Read as | |---|---| | `queries`, `prompts`, `searchQueries`, `questions`, `query`, `prompt`, `keywords`, `searchTerms` | Search queries (a list or a single string) | | `country`, `countryCode`, `gl`, `location`, `region` | Country (code or name) | | `language`, `languageCode`, `lang`, `hl` | Language | | `maxItems`, `maxResults`, `limit`, `maxQueries` | Maximum answers per run | ### One real row Query `best AI coding assistants for developers`, country `US` (collected in the Apify cloud on 2026-10-09; answer and lists shortened here): ```json { "type": "answer", "status": "OK", "query": "best AI coding assistants for developers", "answer": "The choice of the best AI coding assistant depends entirely on your specific workflow, the complexity of your repository, and your team's security requirements. As of **late 2026**, the ecosystem has shifted from simple line completions to advanced **autonomous agents** ...\n\n### The Absolute Top Tier: Best for Day-to-Day Development\n\n- **Cursor**: **Best overall AI-first IDE** for complex, multi-file projects ...\n- **GitHub Copilot**: **Best for seamless editor integration** and standard pair programming ...", "answerText": "The choice of the best AI coding assistant depends entirely on your specific workflow ...", "answerLength": 3597, "sourceCount": 5, "citedSourceCount": 5, "sources": [ { "position": 1, "url": "https://www.shakudo.io/blog/best-ai-coding-assistants", "title": "Best AI Coding Assistants as of October 2026 | Shakudo Blog", "domain": "shakudo.io", "siteName": "Shakudo", "snippet": "Closed-source assistants: GitHub Copilot. Devin (autonomous software engineer); Cursor ...", "citedInAnswer": true }, { "position": 3, "url": "https://www.faros.ai/blog/best-ai-coding-agents-2026", "title": "Best AI Coding Agents for 2026: Real", "domain": "faros.ai", "siteName": "Faros AI", "snippet": "Copilot (Agent Mode): the pragmatic default Cline: VS Code agent for developers who want control ...", "citedInAnswer": true }, { "...": "3 more" } ], "citedDomains": ["shakudo.io", "replit.com", "faros.ai", "zapier.com", "axify.io"], "allDomains": ["shakudo.io", "replit.com", "faros.ai", "zapier.com", "axify.io"], "citationCount": 8, "citations": [ { "position": 3, "text": "GitHub Copilot: Best for seamless editor integration and standard pair programming ...", "sources": [ {"url": "https://www.faros.ai/blog/best-ai-coding-agents-2026", "title": "Best AI Coding Agents for 2026: Real", "domain": "faros.ai"} ] }, { "...": "7 more" } ], "model": "Google AI Mode (logged-out)", "country": "US", "language": "en", "responseSeconds": 18.3, "scrapedAt": "2026-10-09T16:56:32Z", "error": null, "errorCode": null } ``` ### What you can do with it - **AI search visibility (GEO / AEO):** see which brands and domains Google AI Mode recommends and cites for the queries that matter in your market, by country and language. - **Brand monitoring:** add your brand and your competitors to `brands` and get, for every answer, whether each one was mentioned, in what order, and whether its site was cited. - **Content and SEO research:** find the pages Google trusts enough to cite in AI Mode, compare them with the classic results, and see which passages they are credited for. - **Product and market research:** collect how AI Mode compares products, prices and features in a category, in Markdown that keeps the tables and lists. - **Datasets for analysis:** build a dated record of AI answers and citations, run after run, to track how they change. ### Input | Field | What it does | |---|---| | `queries` | Search queries, up to 300 characters each, one answer per query | | `maxItems` | Maximum answers in this run. Default 50 | | `country` | Country Google localizes the answer for. Default `US` | | `language` | Answer and interface language. Empty follows the country | | `brands` | Optional: brands or domains to check in each answer (`Cursor \| cursor.com`) | ```json { "queries": ["best AI coding assistants for developers", "is the iPhone 17 worth upgrading from the iPhone 15"], "country": "US", "maxItems": 50 } ``` With no input at all, the Actor runs one example query, so a test run always returns data. This Actor is built for search queries. Content generation requests (write, generate, translate, code...) and queries over 300 characters are returned as free `INVALID_INPUT` rows. Queries are asked one at a time, each in a fresh AI Mode search, with a short pause between them. #### Brand check example ```json { "queries": ["best AI coding assistants for developers", "best AI code review tools"], "brands": ["Cursor | cursor.com", "GitHub Copilot | github.com", "Windsurf"] } ``` Each row then carries `brands` (one entry per brand with `mentioned`, `mentionCount`, `firstMentionRank`, `cited`, `citationPosition`), `brandsMentioned` and `brandsCited`. The dataset also has a ready "Brand check" view. ### Output fields `type`, `status`, `query`, `answer` (Markdown), `answerText` (plain text), `answerLength`, `sourceCount`, `citedSourceCount`, `sources` (`position`, `url`, `title`, `domain`, `siteName`, `snippet`, `citedInAnswer`), `citedDomains`, `allDomains`, `citationCount`, `citations` (`position`, `text`, `sources`), `model`, `country`, `language`, `responseSeconds`, `scrapedAt`, `error`, `errorCode`. The dataset has three views: **Google AI Mode answers** (one line per query), **Sources** (one line per source) and **Brand check**. Export as JSON, CSV, Excel or through the API. #### Status and error codes (errors never charged) | status / errorCode | Meaning | |---|---| | `OK` | Answer delivered and charged | | `INVALID_INPUT` | Query over 300 characters, or a content generation request | | `NO_ANSWER` | Google loaded but did not show an AI Mode answer for this query | | `BLOCKED` | Google did not answer after several retries; run again later | | `NOT_REACHED`, `UNEXPECTED` | The run timeout arrived, or something unforeseen broke; only that query is lost | ### Pricing | Event | Free plan | Gold and above | |---|---|---| | Google AI Mode answer delivered | US$ 0.035 | US$ 0.02975 | | Run started (once per run that asks at least one valid query) | US$ 0.005 | US$ 0.00425 | 100 queries in one run cost US$ 3.505 on the Free plan. Error rows are free, a run with only invalid queries pays nothing, and the proxy is included. ### Use it from code, Make, Zapier or an AI agent Call it from the Apify API, the Python or JavaScript client, Make, Zapier, n8n or any MCP client. Example with the Python client: ```python from apify_client import ApifyClient client = ApifyClient("") run = client.actor("scrapewise/google-ai-mode-scraper").call(run_input={ "queries": ["best AI coding assistants for developers"], "country": "US", }) for row in client.dataset(run["defaultDatasetId"]).iterate_items(): print(row["query"], row["sourceCount"], row["citedDomains"]) ``` Schedule the same input daily or weekly to follow how the answers and the cited sites change. ### Limitations - Answers are what a logged-out visitor sees in Google AI Mode. AI answers vary from one run to the next; compare several runs before drawing conclusions. - Each answer takes about 30 seconds, one query at a time, so 50 queries take around 25 to 30 minutes. Keep the default run timeout or raise it for long lists; queries the timeout does not reach come back as free `NOT_REACHED` rows. - AI Mode is not offered for every query, country or language. When Google shows no AI answer, the row comes back as a free `NO_ANSWER`. - Country and language localize the answer through Google's own region and language settings. - This Actor reads public answers only and respects the sites' terms. ### FAQ **Do I need a Google account or an API key?** No. The Actor reads AI Mode as a logged-out visitor. **Is it the same as the AI Overview?** No. AI Overview is the short summary on top of normal results; AI Mode is the full conversational answer in its own tab, longer and with more sources. **Can I get the answer in another language?** Yes. Set `language` (for example `de`) and write the query in that language. ### Changelog - **2026-10-09, 0.1:** first release. Google AI Mode answers in Markdown and plain text, cited sources with title and snippet, passage-level citations, any country and language, optional brand check, input names of other AI Mode actors accepted, free error rows. # Actor input Schema ## `queries` (type: `array`): What to ask Google AI Mode, one query per line, up to 300 characters each (for example: best AI coding assistants for developers, is the iPhone 17 worth upgrading from the iPhone 15, Notion vs Obsidian for personal notes). Each query is asked once, in a fresh AI Mode search, and returns one row with the answer and its sources. Content generation requests (write, generate, translate, code...) come back as free INVALID_INPUT rows. Also accepted under the names prompts, searchQueries, questions, query, prompt, keywords and searchTerms. ## `maxItems` (type: `integer`): Upper limit of queries asked in this run (one answer per query). Queries beyond the limit are skipped and never charged. Each answer takes about 30 seconds, so 50 answers finish in roughly 25 to 30 minutes. Raise it (up to 1,000) for larger keyword lists. ## `country` (type: `string`): Two-letter country code Google localizes the answer for (US, GB, CA, AU, DE, FR, ES, IT, NL, MX, IN, JP...). Sent to Google as the search region, so local brands, stores and sources show up. A country name (Germany, Japan) also works; anything unrecognized falls back to US. ## `language` (type: `string`): Interface and answer language (en, es, fr, de, ja, or a full tag like en-GB). Leave empty to follow the country. Google AI Mode usually answers in the language of the query, so write the query in the language you want back. ## `brands` (type: `array`): Optional. A name (Cursor), a domain (cursor.com) or both separated by | (Cursor | cursor.com). Each answer then also says whether the brand was mentioned, how often, its rank among the brands you listed, and whether its domain was cited as a source. Same price, useful for AI visibility and GEO tracking. ## `proxyConfiguration` (type: `object`): Keep Apify Proxy on with the default setting (included in the price). The answer is localized through the Country field, so you do not need a proxy in the target country. ## Actor input object example ```json { "queries": [ "best AI coding assistants for developers", "is the iPhone 17 worth upgrading from the iPhone 15" ], "maxItems": 50, "country": "US", "proxyConfiguration": { "useApifyProxy": true } } ``` # Actor output Schema ## `results` (type: `string`): No description ## `resultsCsv` (type: `string`): No description # API You can run this Actor programmatically using our API. Below are code examples in JavaScript, Python, and CLI, as well as the OpenAPI specification and MCP server setup. ## JavaScript example ```javascript import { ApifyClient } from 'apify-client'; // Initialize the ApifyClient with your Apify API token // Replace the '' with your token const client = new ApifyClient({ token: '', }); // Prepare Actor input const input = { "queries": [ "best AI coding assistants for developers", "is the iPhone 17 worth upgrading from the iPhone 15" ], "proxyConfiguration": { "useApifyProxy": true } }; // Run the Actor and wait for it to finish const run = await client.actor("scrapewise/google-ai-mode-scraper").call(input); // Fetch and print Actor results from the run's dataset (if any) console.log('Results from dataset'); console.log(`💾 Check your data here: https://console.apify.com/storage/datasets/${run.defaultDatasetId}`); const { items } = await client.dataset(run.defaultDatasetId).listItems(); items.forEach((item) => { console.dir(item); }); // 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/js/docs ``` ## Python example ```python from apify_client import ApifyClient # Initialize the ApifyClient with your Apify API token # Replace '' with your token. client = ApifyClient("") # Prepare the Actor input run_input = { "queries": [ "best AI coding assistants for developers", "is the iPhone 17 worth upgrading from the iPhone 15", ], "proxyConfiguration": { "useApifyProxy": True }, } # Run the Actor and wait for it to finish run = client.actor("scrapewise/google-ai-mode-scraper").call(run_input=run_input) # Fetch and print Actor results from the run's dataset (if there are any) print(f"💾 Check your data here: https://console.apify.com/storage/datasets/{run.default_dataset_id}") for item in client.dataset(run.default_dataset_id).iterate_items(): print(item) # 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/python/docs/quick-start ``` ## CLI example ```bash echo '{ "queries": [ "best AI coding assistants for developers", "is the iPhone 17 worth upgrading from the iPhone 15" ], "proxyConfiguration": { "useApifyProxy": true } }' | apify call scrapewise/google-ai-mode-scraper --silent --output-dataset ``` ## MCP server setup ```json { "mcpServers": { "apify": { "type": "http", "url": "https://mcp.apify.com/?tools=fetch-actor-details,scrapewise/google-ai-mode-scraper" } } } ``` The hosted server signs you in with OAuth on first connect, so no API token belongs in this config. Clients without OAuth support can send an `Authorization: Bearer ` header instead, using a token from API & Integrations in Apify Console (https://console.apify.com/settings/integrations). ## OpenAPI specification Download the OpenAPI definition: https://api.apify.com/v2/actors/SfLVxJSyJNaOTzWuK/builds/PEBYKrfskibAPCcZx/openapi.json