{
  "markdown": "# 🔎 Google AI Overview API: Get Google's AI answer and its sources as JSON\n\n> The most efficient, reliable, and developer-friendly way to use the Google AI Overview API.\n\n**Actor page:** [apify.com/johnvc/Google-AI-Overview-API](https://apify.com/johnvc/Google-AI-Overview-API?fpr=9n7kx3)\n**Input schema:** [apify.com/johnvc/Google-AI-Overview-API/input-schema](https://apify.com/johnvc/Google-AI-Overview-API/input-schema?fpr=9n7kx3)\n\nSend any search query and get back Google's AI Overview, the AI-generated answer that appears at the top of Google's results, as structured JSON: the answer text blocks plus the sources it cites (title, link, snippet, and source name). When Google defers generation and returns only a short-lived token, this API resolves the full answer for you automatically, so you always get clean, ready-to-use data.\n\n## Video Walkthrough\n\n[![Watch the walkthrough](https://img.youtube.com/vi/jREWahDGhJM/maxresdefault.jpg)](https://www.youtube.com/watch?v=jREWahDGhJM)\n\n## Quick Start\n\n### Prerequisites\n- Python 3.11 or higher\n- An Apify account and API key ([get a free key here](https://apify.com?fpr=9n7kx3))\n\n1. **Clone the repository**\n   ```bash\n   git clone https://github.com/johnisanerd/Apify-Google-AI-Overview-API.git\n   cd Apify-Google-AI-Overview-API\n   ```\n\n2. **Install dependencies with UV**\n   ```bash\n   # Install UV if you do not have it:\n   curl -LsSf https://astral.sh/uv/install.sh | sh\n\n   # Install project dependencies:\n   uv sync\n   ```\n\n3. **Configure your API key**\n   ```bash\n   cp .env.example .env\n   # Edit .env and add your Apify API key\n   # Get your free API key at: https://apify.com?fpr=9n7kx3\n   ```\n\n4. **Run the example**\n   ```bash\n   uv run python google-ai-overview-api-example.py\n   ```\n\n### Alternative: set the API key directly\n```bash\nexport APIFY_API_TOKEN=\"your_api_key_here\"\nuv run python google-ai-overview-api-example.py\n```\n\n## Why Use This Google AI Overview API?\n\n**Just the answer, structured.** You get the AI Overview content and its cited sources as clean JSON, ready to drop into a database, a report, or an LLM pipeline. No HTML parsing.\n\n**It handles Google's deferral for you.** Google sometimes returns the AI Overview immediately and sometimes defers it behind a short-lived token. This API resolves that follow-up automatically in a single run, so you never have to manage tokens that expire in under a minute.\n\n**One query or many.** Pass a single `query` or a `queries` list. You get one result row per query, which makes batch lookups and GEO/AEO monitoring straightforward.\n\n**Pay only for what you fetch.** Pricing is per retrieval: one retrieval for a standard inline answer, two only when Google defers and a follow-up is required. The output tells you exactly how many were used.\n\n**Country and language targeting.** Choose the country (`gl`) and language (`hl`), or narrow to a specific `location`.\n\n**MCP-ready.** Load it as a tool in Claude Cowork, Claude Code, Claude on the web, and Cursor (see below) and ask for an AI Overview in plain language.\n\n## Features\n\n### Core Capabilities\n- Returns the AI Overview answer as ordered `text_blocks` (paragraphs, headings, lists)\n- Returns the cited `references` for each answer (title, link, snippet, source)\n- Automatic resolution of Google's deferred (token-based) generation\n- Single-query and batch (`queries`) input, one row per query\n- Country, language, and location targeting\n\n### Data Quality\n- Clean, typed JSON output, one row per query, with a stable `result_type` field\n- A clear `ai_overview_present` flag and a `note` when Google shows no overview\n- Transparent `retrievals_used` count so cost is never a surprise\n\n## Usage Examples\n\n### Basic Example\n```json\n{\n  \"query\": \"what is the Siliconimist podcast?\"\n}\n```\n\n### Advanced Example\n```json\n{\n  \"queries\": [\n    \"best practices for prompt engineering\",\n    \"how does vector search work\"\n  ],\n  \"gl\": \"us\",\n  \"hl\": \"en\",\n  \"location\": \"Austin, Texas, United States\"\n}\n```\n\n## Input Parameters\n\n| Parameter | Type | Required | Default | Description |\n|-----------|------|----------|---------|-------------|\n| `query` | `str` | one of `query` / `queries` | - | A single search query to fetch the AI Overview for. |\n| `queries` | `list[str]` | one of `query` / `queries` | - | Multiple queries in one run; one result row per query. Merged with `query`. |\n| `gl` | `str` | no | `us` | Country code (ISO 3166-1), e.g. `us`, `gb`, `ca`. |\n| `hl` | `str` | no | `en` | Language code (ISO 639-1). AI Overviews are currently English-only. |\n| `location` | `str` | no | - | Optional location to narrow results, e.g. `Austin, Texas, United States`. |\n\nAt least one of `query` or `queries` is required.\n\n## Output Format\n\nOne row per query. Each item has a `result_type` of `ai_overview` or `error`. The answer text and sources below are illustrative; your run returns the live AI Overview content and its real citations.\n\n```json\n{\n  \"result_type\": \"ai_overview\",\n  \"query\": \"what is retrieval augmented generation\",\n  \"ai_overview_present\": true,\n  \"used_dedicated_api\": false,\n  \"retrievals_used\": 1,\n  \"text_blocks\": [\n    {\n      \"type\": \"paragraph\",\n      \"snippet\": \"<AI Overview answer paragraph for the query>\"\n    }\n  ],\n  \"references\": [\n    {\n      \"title\": \"<cited source title>\",\n      \"link\": \"https://www.example.com/source\",\n      \"snippet\": \"<short snippet from the cited source>\",\n      \"source\": \"Example\",\n      \"index\": 1\n    }\n  ],\n  \"gl\": \"us\",\n  \"hl\": \"en\",\n  \"fetched_at\": \"2026-05-28T08:33:58.988871+00:00\"\n}\n```\n\nWhen Google does not show an AI Overview for a query, the row comes back with `\"ai_overview_present\": false` and a short `note` explaining why.\n\n---\n\n## Use as an MCP tool\n\nYou can load the Google AI Overview API as an MCP tool so assistants call it for you. The MCP server URL preloads just this one Actor:\n\n```\nhttps://mcp.apify.com/?tools=actors,docs,johnvc/Google-AI-Overview-API\n```\n\nAuthenticate with OAuth in the browser when offered, or with your Apify API token (the same `APIFY_API_TOKEN` used by the Python example). Get a token at https://console.apify.com/settings/integrations and a free Apify account at https://apify.com?fpr=9n7kx3 .\n\n## Install in Claude Cowork Desktop\n\n![Install in Claude Cowork Desktop](https://raw.githubusercontent.com/johnisanerd/ApifyPublicData/main/assets/guides/install_mcp_into_claude_desktop.png)\n\nCowork is the desktop app's automation mode. To give it the Google AI Overview API as a tool, add the Apify MCP server as a connector.\n\n1. Open the Claude desktop app and go to **Settings → Connectors** (or **Settings → Developer → Edit Config** to edit `claude_desktop_config.json` directly).\n   - macOS: `~/Library/Application Support/Claude/claude_desktop_config.json`\n   - Windows: `%APPDATA%\\Claude\\claude_desktop_config.json`\n2. Add the Apify MCP server, preloaded with only this Actor:\n\n```json\n{\n  \"mcpServers\": {\n    \"apify\": {\n      \"command\": \"npx\",\n      \"args\": [\n        \"-y\",\n        \"mcp-remote\",\n        \"https://mcp.apify.com/?tools=actors,docs,johnvc/Google-AI-Overview-API\"\n      ]\n    }\n  }\n}\n```\n\n3. Restart the app. When Cowork first calls the tool, complete the OAuth prompt in your browser, or add your Apify API token in the connector settings to skip OAuth.\n4. In a Cowork chat, confirm the tool is available and ask it to run the Google AI Overview API.\n\nDownload the desktop app and start a free trial: https://claude.ai/referral/uIlpa7nPLg\nMore help: https://docs.apify.com/platform/integrations/claude-desktop\n\n## Install in Claude Code\n\n![Install in Claude Code](https://raw.githubusercontent.com/johnisanerd/ApifyPublicData/main/assets/guides/install_mcp_into_claude_code.png)\n\nClaude Code is the command-line tool. Add the Actor's MCP server with one command:\n\n```bash\nclaude mcp add --transport http apify \\\n  \"https://mcp.apify.com/?tools=actors,docs,johnvc/Google-AI-Overview-API\"\n```\n\nTo use a token instead of browser OAuth:\n\n```bash\nclaude mcp add --transport http apify \\\n  \"https://mcp.apify.com/?tools=actors,docs,johnvc/Google-AI-Overview-API\" \\\n  --header \"Authorization: Bearer YOUR_APIFY_TOKEN\"\n```\n\nThen verify with `claude mcp list`, or run `/mcp` inside a session. Ask Claude Code to call the Google AI Overview API.\n\nTry Claude Code free: https://claude.ai/referral/uIlpa7nPLg\nClaude Code MCP docs: https://code.claude.com/docs/en/mcp\n\n## Install in Claude (website)\n\n![Install in Claude (website)](https://raw.githubusercontent.com/johnisanerd/ApifyPublicData/main/assets/guides/install_mcp_into_claude_ai.png)\n\nOn claude.ai you add Apify as a connector, then enable just this Actor's tool.\n\n1. Go to **Settings → Connectors → Browse connectors** and search for **Apify MCP server**. Install it (enable or update if prompted).\n2. When connecting, authenticate with your Apify API token, and enable the tool `johnvc/Google-AI-Overview-API`.\n3. In any chat, open **+ → Connectors** and turn on **Apify**.\n4. Alternatively, choose **Add custom connector** and paste the full MCP URL `https://mcp.apify.com/?tools=actors,docs,johnvc/Google-AI-Overview-API`, using OAuth when prompted.\n5. Ask Claude to run the Google AI Overview API.\n\nOpen Claude on the web: https://claude.ai/referral/uIlpa7nPLg\n\n## Install in Cursor\n\n![Install in Cursor](https://raw.githubusercontent.com/johnisanerd/ApifyPublicData/main/assets/guides/install_mcp_into_cursor.png)\n\nCursor reads MCP servers from a project file at `.cursor/mcp.json`.\n\n1. In your project, create `.cursor/mcp.json`:\n\n```json\n{\n  \"mcpServers\": {\n    \"apify\": {\n      \"url\": \"https://mcp.apify.com/?tools=actors,docs,johnvc/Google-AI-Overview-API\"\n    }\n  }\n}\n```\n\n2. If you prefer token auth over browser OAuth, add a header:\n\n```json\n{\n  \"mcpServers\": {\n    \"apify\": {\n      \"url\": \"https://mcp.apify.com/?tools=actors,docs,johnvc/Google-AI-Overview-API\",\n      \"headers\": { \"Authorization\": \"Bearer YOUR_APIFY_TOKEN\" }\n    }\n  }\n}\n```\n\n3. Open **Cursor → Settings → MCP** and confirm the **apify** server is connected (green dot).\n4. In Composer or Chat, ask Cursor to call the Google AI Overview API.\n\nNew to Cursor? Get it here: https://cursor.com/referral?code=XQP4VBLI3NNX\n\n## Install in ChatGPT\n\n![Install in ChatGPT](https://raw.githubusercontent.com/johnisanerd/ApifyPublicData/main/assets/guides/install_mcp_into_ChatGPT.png)\n\nChatGPT connects to the Apify MCP server through Developer mode (available on ChatGPT Pro, Plus, Business, Enterprise, and Education plans).\n\n1. Click your profile icon, then go to **Settings > Apps**. If you do not see a **Create app** button, open **Advanced settings** and enable **Developer mode**.\n2. Click **Create app** and fill out the form:\n   - **Name:** Apify\n   - **MCP Server URL:** `https://mcp.apify.com/?tools=actors,docs,johnvc/Google-AI-Overview-API`\n   - **Authentication:** OAuth\n3. Click **Create** and authorize the connection with Apify.\n4. To use the app in a conversation, click **+** in the chat, choose **Developer mode**, and select **Apify**.\n\nMore help: https://docs.apify.com/platform/integrations/mcp\n\n---\n\n[**Made with care**](https://apify.com/johnvc?fpr=9n7kx3)\n\n*Use the Google AI Overview API to power GEO/AEO monitoring, research, and content workflows with reliable, structured results.*\n\nLast Updated: 2026.09.07\n",
  "bytes": 11278,
  "sha": "f901fca9ee08fb21069a5fdac27511598ff7018b6c38842477aad11e63e1eb91",
  "repo_slug": "johnisanerd/apify-google-ai-overview-api",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_johnisanerd_google_ai_overview_ce5012a2/readme"
}