> For the complete documentation index, see [llms.txt](https://docs.agenticflow.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.agenticflow.ai/reference/nodes/google_search.md).

# Google Search

**Action ID:** `google_search`

## Description

Search Google for recent results.

## Connection

| Name               | Description                                 | Required | Category |
| ------------------ | ------------------------------------------- | -------- | -------- |
| PixelML Connection | The PixelML connection to call PixelML API. | True     | pixelml  |

## Input Parameters

| Name          | Type    | Required | Default | Description                                   |
| ------------- | ------- | :------: | ------- | --------------------------------------------- |
| search\_query | string  |     ✓    | -       | What do you want to search for?               |
| top\_k        | integer |     -    | 5       | Number of top search results to return (1-10) |

<details>

<summary>View JSON Schema</summary>

```json
{
  "description": "GoogleSearch node runner input.",
  "properties": {
    "search_query": {
      "description": "What do you want to search for?",
      "title": "Search Query",
      "type": "string"
    },
    "top_k": {
      "default": 5,
      "description": "Number of top search results to return",
      "maximum": 10,
      "minimum": 1,
      "title": "Top k",
      "type": "integer"
    }
  },
  "required": [
    "search_query"
  ],
  "title": "GoogleSearchNodeRunnerInput",
  "type": "object"
}
```

</details>

## Output Parameters

| Name            | Type  | Description                                                  |
| --------------- | ----- | ------------------------------------------------------------ |
| search\_results | array | Array of search result objects with title, snippet, and link |

<details>

<summary>View JSON Schema</summary>

```json
{
  "$defs": {
    "GoogleSearchResult": {
      "description": "Google search result.",
      "properties": {
        "title": {
          "title": "Title",
          "type": "string"
        },
        "snippet": {
          "title": "Snippet",
          "type": "string"
        },
        "link": {
          "title": "Link",
          "type": "string"
        }
      },
      "required": [
        "title",
        "snippet",
        "link"
      ],
      "title": "GoogleSearchResult",
      "type": "object"
    }
  },
  "description": "GoogleSearch node runner output.",
  "properties": {
    "search_results": {
      "description": "Top k search results",
      "items": {
        "$ref": "#/$defs/GoogleSearchResult"
      },
      "title": "Search Results",
      "type": "array"
    }
  },
  "required": [
    "search_results"
  ],
  "title": "GoogleSearchNodeRunnerOutput",
  "type": "object"
}
```

</details>

## How It Works

This node performs Google searches through the PixelML API and returns structured search results. You provide a search query, and the node retrieves the top matching results including title, snippet, and URL. The top\_k parameter controls how many results are returned (1-10). Results are presented in relevance order as determined by Google's algorithm.

## Usage Examples

### Example 1: Search for Product Information

**Input:**

```
search_query: "best electric cars 2024"
top_k: 5
```

**Output:**

```
search_results: [
  {
    "title": "Top 10 Electric Cars of 2024 - Auto Review",
    "snippet": "Comprehensive review of the best electric vehicles available in 2024...",
    "link": "https://autoreview.com/top-electric-cars-2024"
  },
  {
    "title": "2024 Electric Car Buyer's Guide",
    "snippet": "Everything you need to know about buying an electric car this year...",
    "link": "https://carguide.com/electric-2024"
  }
]
```

### Example 2: Search for Recent News

**Input:**

```
search_query: "latest AI developments"
top_k: 3
```

**Output:**

```
search_results: [
  {
    "title": "Major AI Breakthrough Announced Today",
    "snippet": "Researchers unveil new language model with enhanced capabilities...",
    "link": "https://technews.com/ai-breakthrough"
  }
]
```

### Example 3: Research Competitor Information

**Input:**

```
search_query: "competitor company latest product launch"
top_k: 10
```

**Output:**

```
search_results: [... 10 search results ...]
```

## Common Use Cases

* **Market Research**: Gather information about competitors, products, or market trends
* **Content Research**: Find sources and references for content creation
* **News Monitoring**: Track latest news and updates on specific topics
* **SEO Analysis**: Understand what content ranks for specific keywords
* **Product Research**: Find product reviews, comparisons, and specifications
* **Trend Analysis**: Identify trending topics and discussions in your industry
* **Fact Checking**: Verify information by searching for authoritative sources

## Error Handling

| Error Type           | Cause                                         | Solution                                        |
| -------------------- | --------------------------------------------- | ----------------------------------------------- |
| Connection Failed    | Invalid or expired PixelML connection         | Verify your PixelML connection credentials      |
| Empty Query          | Search query is empty or null                 | Provide a non-empty search query string         |
| No Results Found     | Query returned no matching results            | Try different search terms or broader keywords  |
| Rate Limit Exceeded  | Too many search requests in short period      | Space out requests or upgrade your PixelML plan |
| Invalid top\_k Value | top\_k is outside allowed range (1-10)        | Set top\_k between 1 and 10                     |
| Service Unavailable  | Google search service temporarily unavailable | Retry the request after a short delay           |
| Query Too Complex    | Search query is too long or complex           | Simplify the search query                       |

## Notes

* **Result Limit**: Maximum 10 results per search. For more comprehensive research, perform multiple searches with different queries.
* **Search Freshness**: Results reflect Google's current index. Very recent content may not appear immediately.
* **Cost Per Search**: Each search operation consumes PixelML API credits. Monitor usage to control costs.
* **Query Syntax**: Use natural language queries. Advanced Google search operators may not be supported.
* **Result Structure**: Each result includes title, snippet (description), and full URL.
* **Rate Limiting**: PixelML implements rate limiting. Space out searches appropriately.
* **Regional Results**: Results may vary by region. The node uses PixelML's default regional settings.
* **Link Validity**: URLs are returned as-is. Verify links are accessible before using them in production.