> **Building with AI coding agents?** Install the authstack plugin with one command. This equips your agent with accurate Scalekit implementation patterns.
>
> **Recommended** (the `-y` flags skip prompts, so the command runs without an interactive terminal):
> ```bash
> npx -y @scalekit-inc/cli setup -y
> ```
>
> Global:
> ```bash
> npm install -g @scalekit-inc/cli
> scalekit setup -y
> ```
>
> Supports Claude Code, Cursor, GitHub Copilot, Codex + skills for other Agent Skills-compatible agents in `~/.agents/skills`.
> Skills: integrate-agentkit, implement-saaskit, add-mcp-oauth, implement-sso, implement-scim.
> [Full setup guide](https://docs.scalekit.com/dev-kit/build-with-ai/)

---

# Connect AI agents to SerpApi

The SerpApi connector lets your AI agent act in each user's SerpApi account. Each user connects their own SerpApi API key once, and Scalekit sends it with every call, so your agent never handles credentials. It comes with 19 tools.

**Authentication:** API Key
**Categories:** Search, Marketing
**Tools:** 19: 19 read, 0 write, 0 destructive
**Users sign in with:** API key
**Built by:** Scalekit connector
**Try it:** [Playground in the Scalekit dashboard](https://app.scalekit.com/ws/signup?sk_intent=playground&provider=SERPAPI)

## Setup

1. ### Install the SDK

   **Node.js**

   ```bash
   npm install @scalekit-sdk/node dotenv
   ```

   **Python**

   ```bash
   pip install scalekit-sdk-python python-dotenv
   ```

2. ### Set your credentials

   Add your Scalekit credentials to your `.env` file. Find values in **[app.scalekit.com](https://app.scalekit.com)** > **Developers** > **API Credentials**.

   ```sh title=".env"
   SCALEKIT_ENVIRONMENT_URL=<your-environment-url>
   SCALEKIT_CLIENT_ID=<your-client-id>
   SCALEKIT_CLIENT_SECRET=<your-client-secret>
   ```

3. ### Create the SerpApi connection

   In **AgentKit > Connections**, create a SerpApi connection. The name you give it is the `connection_name` your code passes. See [Configure connections](/agentkit/connections/).

4. ### Authorize a user and make your first call

   **Node.js** (`quickstart.mts`)

   ```typescript
   import { ScalekitClient } from '@scalekit-sdk/node'
   import 'dotenv/config'
   import { createInterface } from 'node:readline/promises'

   const scalekit = new ScalekitClient(
     process.env.SCALEKIT_ENVIRONMENT_URL,
     process.env.SCALEKIT_CLIENT_ID,
     process.env.SCALEKIT_CLIENT_SECRET,
   )
   const actions = scalekit.actions

   const connector = 'serpapi'
   const identifier = 'user_123'

   // Generate an authorization link for the user
   const { link } = await actions.getAuthorizationLink({ connectionName: connector, identifier })
   console.log('Authorize SerpApi:', link)
   const rl = createInterface({ input: process.stdin, output: process.stdout })
   await rl.question('Press Enter after authorizing...')
   rl.close()

   // Make your first call
   const result = await actions.executeTool({
     connector,
     identifier,
     toolName: 'serpapi_account_get',
     toolInput: {},
   })
   console.log(result)
   ```

   ```bash
   npx tsx quickstart.mts
   ```

   **Python** (`quickstart.py`)

   ```python
   import os
   from scalekit import ScalekitClient
   from dotenv import load_dotenv
   load_dotenv()

   scalekit_client = ScalekitClient(
       env_url=os.getenv("SCALEKIT_ENVIRONMENT_URL"),
       client_id=os.getenv("SCALEKIT_CLIENT_ID"),
       client_secret=os.getenv("SCALEKIT_CLIENT_SECRET"),
   )
   actions = scalekit_client.actions

   connection_name = "serpapi"
   identifier = "user_123"

   # Generate an authorization link for the user
   link_response = actions.get_authorization_link(
       connection_name=connection_name,
       identifier=identifier,
   )
   print("Authorize SerpApi:", link_response.link)
   input("Press Enter after authorizing...")

   # Make your first call
   result = actions.execute_tool(
       tool_input={},
       tool_name="serpapi_account_get",
       connection_name=connection_name,
       identifier=identifier,
   )
   print(result)
   ```

   ```bash
   python quickstart.py
   ```

   Each user opens the link once and enters their SerpApi credentials there. See [Authorize a user](/agentkit/tools/authorize/) for the full flow and statuses.

## Tools

Pass the exact name to `execute_tool`, with an input like each tool's example. To ask for a tool that's missing, use the [request form](https://scalekitsupport.portal.usepylon.com/forms/request-a-connector-tool).

**Node.js**

```typescript
const result = await actions.executeTool({
  toolName: 'serpapi_account_get',
  toolInput: {},
  connector: 'serpapi',
  identifier: 'user_123',
})
```

**Python**

```python
result = actions.execute_tool(
    tool_name="serpapi_account_get",
    tool_input={},
    connection_name="serpapi",
    identifier="user_123",
)
```

`result.data` is the app's response as JSON, and `result.execution_id` (`executionId` in Node.js) is the ID of the call.

### `serpapi_account_get`

Get Account · Read-only

Get the SerpApi account's plan, remaining searches and usage. This call is free.
Returns account_email, account_status, plan_name, searches_per_month, plan_searches_left, extra_credits, total_searches_left, this_month_usage, this_hour_searches, last_hour_searches and account_rate_limit_per_hour.
Use this before a large batch of searches to check the remaining balance and hourly limit.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `serpapi_amazon_search`

Amazon Search · Read-only

Search Amazon product listings through SerpApi, by keyword or by category node.
Returns organic_results with position, asin, title, link, price, extracted_price, rating, reviews, prime, delivery and thumbnail, plus product_ads, featured_products, related_searches, filters and serpapi_pagination.
Use this for Amazon-only product results. Use serpapi_google_shopping_search to compare prices across many stores.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `amazon_domain` (`string`, optional): Amazon site to search. SerpApi uses amazon.com when omitted. Example: amazon.co.uk
- `async` (`boolean`, optional): Set true to submit the search and return at once with only search_metadata (id and status Processing); fetch the results later with serpapi_search_archive_get using search_metadata.id. Do not combine with no_cache. Example: false
- `dc` (`boolean`, optional): Set true to exclude results for an auto-corrected spelling of the query (default false). Example: true
- `delivery_zip` (`string`, optional): ZIP or postal code to show delivery and availability for. Not valid on amazon.ae, amazon.sa or amazon.eg. Example: 10001
- `json_restrictor` (`string`, optional): Limit the response to the listed fields to keep it small, using SerpApi's JSON Restrictor syntax: comma-separated top-level keys, with dots, brackets and braces for nested fields. Example: organic_results[0:5].{title,link,snippet}
- `k` (`string`, optional): The Amazon search query. Required unless node is set; cannot be combined with node. Example: usb c charger 65w
- `language` (`string`, optional): Language as language_REGION. Example: en_US
- `no_cache` (`boolean`, optional): Set true to force a fresh search instead of a cached result. SerpApi serves an identical search run in the last hour from its cache for free; a fresh search spends one search credit. Do not combine with async. Example: false
- `node` (`string`, optional): Amazon category node ID to browse instead of a keyword search. Cannot be combined with k. Example: 172282
- `page` (`integer`, optional): Results page, starting at 1. Each page is a separate search. Example: 2
- `rh` (`string`, optional): Attribute filters as comma-separated key:value pairs, taken from filters in an earlier result. Example: n:16318031,p_72:1248897011
- `s` (`string`, optional): Sort order: relevanceblender featured (default), price-asc-rank, price-desc-rank, review-rank (average review), date-desc-rank (newest), exact-aware-popularity-rank (best sellers). One of: `relevanceblender`, `price-asc-rank`, `price-desc-rank`, `review-rank`, `date-desc-rank`, `exact-aware-popularity-rank`.

Example input: `{}`

### `serpapi_bing_news_search`

Bing News Search · Read-only

Search Bing News through SerpApi, optionally limited to a recent time window or sorted by date.
Returns organic_results with title, link, snippet, source, date and thumbnail.
Use this for Bing's news coverage of a topic. Use serpapi_google_news_search for Google News.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `q` (`string`, required): The news search query. Bing operators such as site: work. Example: electric vehicle battery recycling
- `async` (`boolean`, optional): Set true to submit the search and return at once with only search_metadata (id and status Processing); fetch the results later with serpapi_search_archive_get using search_metadata.id. Do not combine with no_cache. Example: false
- `cc` (`string`, optional): Two-letter country code to search from, such as us or de. Cannot be combined with mkt. Example: gb
- `count` (`integer`, optional): Suggested number of results per page; Bing may return a different number. Example: 20
- `first` (`integer`, optional): Position of the first result to return, for pagination (default 1). Example: 11
- `json_restrictor` (`string`, optional): Limit the response to the listed fields to keep it small, using SerpApi's JSON Restrictor syntax: comma-separated top-level keys, with dots, brackets and braces for nested fields. Example: organic_results[0:5].{title,link,snippet}
- `mkt` (`string`, optional): Bing market as language-country, such as en-US or de-DE. Cannot be combined with cc. Example: en-US
- `no_cache` (`boolean`, optional): Set true to force a fresh search instead of a cached result. SerpApi serves an identical search run in the last hour from its cache for free; a fresh search spends one search credit. Do not combine with async. Example: false
- `qft` (`string`, optional): Time filter or sort: interval="4" past hour, interval="7" past 24 hours, interval="8" past 7 days, interval="9" past 30 days, sortbydate="1" newest first. Omit for best match. One of: `interval="4"`, `interval="7"`, `interval="8"`, `interval="9"`, `sortbydate="1"`.
- `safeSearch` (`string`, optional): Adult content filter: Off, Moderate (adult text allowed, no adult images or videos) or Strict. One of: `Off`, `Moderate`, `Strict`.

Example input: `{"q":"<q>"}`

### `serpapi_bing_search`

Bing Search · Read-only

Search Bing web results through SerpApi.
Returns organic_results with position, title, link, displayed_link and snippet, plus ads, answer_box, knowledge_graph, related_questions, related_searches and serpapi_pagination when Bing shows them.
Use this for Bing's web index or to cross-check Google. Use serpapi_google_search for Google results and serpapi_bing_news_search for Bing news.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `q` (`string`, required): The search query. Bing operators such as site:, NOT and OR work. Example: open source vector database
- `async` (`boolean`, optional): Set true to submit the search and return at once with only search_metadata (id and status Processing); fetch the results later with serpapi_search_archive_get using search_metadata.id. Do not combine with no_cache. Example: false
- `cc` (`string`, optional): Two-letter country code to search from, such as us or de. Cannot be combined with mkt. Example: gb
- `device` (`string`, optional): Device type to emulate: desktop (default), tablet or mobile. Results and layout differ by device. One of: `desktop`, `tablet`, `mobile`.
- `filters` (`string`, optional): Bing filter string, such as a date range, copied from the filters parameter of a Bing results URL. Example: ex1:"ez1"
- `first` (`integer`, optional): Position of the first result to return, for pagination (default 1). Example: 11 for the second page.
- `json_restrictor` (`string`, optional): Limit the response to the listed fields to keep it small, using SerpApi's JSON Restrictor syntax: comma-separated top-level keys, with dots, brackets and braces for nested fields. Example: organic_results[0:5].{title,link,snippet}
- `location` (`string`, optional): Place the search runs from, as a location name (city level works best). Example: Austin, Texas, United States
- `mkt` (`string`, optional): Bing market as language-country, such as en-US or de-DE. Cannot be combined with cc. Example: en-US
- `no_cache` (`boolean`, optional): Set true to force a fresh search instead of a cached result. SerpApi serves an identical search run in the last hour from its cache for free; a fresh search spends one search credit. Do not combine with async. Example: false
- `safeSearch` (`string`, optional): Adult content filter: Off, Moderate (adult text allowed, no adult images or videos) or Strict. One of: `Off`, `Moderate`, `Strict`.

Example input: `{"q":"<q>"}`

### `serpapi_google_autocomplete`

Google Autocomplete · Read-only

Get Google's autocomplete suggestions for a partial query through SerpApi.
Returns suggestions with value (the suggested query), relevance and type, plus verbatim_relevance.
Use this to expand keywords or see how people phrase a search. Use serpapi_google_trends_search for popularity over time.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `q` (`string`, required): The partial query to complete. Example: how to make cold br
- `async` (`boolean`, optional): Set true to submit the search and return at once with only search_metadata (id and status Processing); fetch the results later with serpapi_search_archive_get using search_metadata.id. Do not combine with no_cache. Example: false
- `cp` (`integer`, optional): Cursor position in the query, starting at 0; suggestions complete the text at this position. Defaults to the end of the query. Example: 10
- `gl` (`string`, optional): Two-letter country code (Google's gl parameter, lowercase) that sets which country's results to return. Example: us
- `hl` (`string`, optional): Language code for the results (Google's hl parameter): a two-letter language code, optionally with a region, such as en, fr, en-gb or es-419. Example: en
- `json_restrictor` (`string`, optional): Limit the response to the listed fields to keep it small, using SerpApi's JSON Restrictor syntax: comma-separated top-level keys, with dots, brackets and braces for nested fields. Example: organic_results[0:5].{title,link,snippet}
- `no_cache` (`boolean`, optional): Set true to force a fresh search instead of a cached result. SerpApi serves an identical search run in the last hour from its cache for free; a fresh search spends one search credit. Do not combine with async. Example: false

Example input: `{"q":"<q>"}`

### `serpapi_google_images_search`

Google Images Search · Read-only

Search Google Images through SerpApi, with optional size, color, type, license and date filters.
Returns images_results with position, title, source, link (page), original (full-size image URL), original_width, original_height and thumbnail, plus suggested_searches and serpapi_pagination.
Use this to find images on a subject. Use serpapi_google_search for web pages and serpapi_google_shopping_search for product listings.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `q` (`string`, required): The image search query. Operators such as site: and intitle: work. Example: golden gate bridge at night
- `async` (`boolean`, optional): Set true to submit the search and return at once with only search_metadata (id and status Processing); fetch the results later with serpapi_search_archive_get using search_metadata.id. Do not combine with no_cache. Example: false
- `device` (`string`, optional): Device type to emulate: desktop (default), tablet or mobile. Results and layout differ by device. One of: `desktop`, `tablet`, `mobile`.
- `end_date` (`string`, optional): Latest image date, as YYYYMMDD. Cannot be combined with period_unit. Example: 20261031
- `gl` (`string`, optional): Two-letter country code (Google's gl parameter, lowercase) that sets which country's results to return. Example: us
- `google_domain` (`string`, optional): Google domain to search on. SerpApi uses google.com when omitted. Example: google.co.uk
- `hl` (`string`, optional): Language code for the results (Google's hl parameter): a two-letter language code, optionally with a region, such as en, fr, en-gb or es-419. Example: en
- `ijn` (`integer`, optional): Page number of image results, starting at 0 (up to 99). Each page is a separate search. Example: 1
- `image_color` (`string`, optional): Dominant color filter: bw (black and white), trans (transparent), or a color name. One of: `bw`, `trans`, `red`, `orange`, `yellow`, `green`, `teal`, `blue`, `purple`, `pink`, `white`, `gray`, `black`, `brown`.
- `image_type` (`string`, optional): Image type filter: face, photo, clipart, lineart or animated. One of: `face`, `photo`, `clipart`, `lineart`, `animated`.
- `imgar` (`string`, optional): Aspect ratio filter: s square, t tall, w wide, xw panoramic. One of: `s`, `t`, `w`, `xw`.
- `imgsz` (`string`, optional): Size filter: l large, m medium, i icon, or a minimum size such as qsvga, vga, svga, xga, 2mp, 4mp, 6mp, 8mp, 10mp, 12mp, 15mp, 20mp, 40mp or 70mp. Example: l
- `json_restrictor` (`string`, optional): Limit the response to the listed fields to keep it small, using SerpApi's JSON Restrictor syntax: comma-separated top-level keys, with dots, brackets and braces for nested fields. Example: organic_results[0:5].{title,link,snippet}
- `licenses` (`string`, optional): Usage rights filter: f free to use or share, fc free to use or share even commercially, fm free to use, share or modify, fmc free to use, share or modify even commercially, cl Creative Commons licenses, ol commercial and other licenses. One of: `f`, `fc`, `fm`, `fmc`, `cl`, `ol`.
- `location` (`string`, optional): Place the search runs from, as a location name (city level works best, most specific first). Results are localized as if searched there. Find valid names with serpapi_locations_search. Example: Austin, Texas, United States
- `no_cache` (`boolean`, optional): Set true to force a fresh search instead of a cached result. SerpApi serves an identical search run in the last hour from its cache for free; a fresh search spends one search credit. Do not combine with async. Example: false
- `period_unit` (`string`, optional): Only return images from the recent past, in this unit: s second, n minute, h hour, d day, w week, m month, y year. Combine with period_value. Cannot be combined with start_date or end_date. One of: `s`, `n`, `h`, `d`, `w`, `m`, `y`.
- `period_value` (`integer`, optional): How many period_unit units back to look (SerpApi uses 1 when omitted). Example: 2 with period_unit w means the past two weeks.
- `safe` (`string`, optional): Adult content filter: active to filter explicit results, off to turn filtering off. Google blurs explicit content when omitted. One of: `active`, `off`.
- `start_date` (`string`, optional): Earliest image date, as YYYYMMDD. Cannot be combined with period_unit. Example: 20260101

Example input: `{"q":"<q>"}`

### `serpapi_google_jobs_search`

Google Jobs Search · Read-only

Search job listings that Google Jobs aggregates from job boards and company sites, through SerpApi.
Returns jobs_results with title, company_name, location, via (source), description, job_highlights, detected_extensions (posted_at, schedule_type, salary), apply_options and job_id, plus filters and serpapi_pagination.next_page_token (up to 10 jobs per page).
Use this to find open roles by title, skill or company and place.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `q` (`string`, required): The job search query, such as a title, skill or company. Example: senior data engineer
- `async` (`boolean`, optional): Set true to submit the search and return at once with only search_metadata (id and status Processing); fetch the results later with serpapi_search_archive_get using search_metadata.id. Do not combine with no_cache. Example: false
- `gl` (`string`, optional): Two-letter country code (Google's gl parameter, lowercase) that sets which country's results to return. Example: us
- `google_domain` (`string`, optional): Google domain to search on. SerpApi uses google.com when omitted. Example: google.co.uk
- `hl` (`string`, optional): Language code for the results (Google's hl parameter): a two-letter language code, optionally with a region, such as en, fr, en-gb or es-419. Example: en
- `json_restrictor` (`string`, optional): Limit the response to the listed fields to keep it small, using SerpApi's JSON Restrictor syntax: comma-separated top-level keys, with dots, brackets and braces for nested fields. Example: organic_results[0:5].{title,link,snippet}
- `location` (`string`, optional): Place the search runs from, as a location name (city level works best, most specific first). Results are localized as if searched there. Find valid names with serpapi_locations_search. Example: Austin, Texas, United States
- `lrad` (`integer`, optional): Search radius in kilometers around the location. It biases results and is not a strict limit. Example: 25
- `next_page_token` (`string`, optional): Token for the next page, from serpapi_pagination.next_page_token of the previous result. Example: eyJmYyI6IkVvd0RDc3dDUVVGQlFXRXRVa0Z
- `no_cache` (`boolean`, optional): Set true to force a fresh search instead of a cached result. SerpApi serves an identical search run in the last hour from its cache for free; a fresh search spends one search credit. Do not combine with async. Example: false
- `uds` (`string`, optional): Filter string from a filters entry of an earlier result (for example a date posted or job type filter). Example: ADvngMjcH0KdF7qJ

Example input: `{"q":"<q>"}`

### `serpapi_google_local_search`

Google Local Search · Read-only

Search Google's local results (the Places list behind the local pack) through SerpApi.
Returns local_results with position, title, place_id, rating, reviews, price, type, address, phone, hours, gps_coordinates and thumbnail, plus local_map.
Use this for a ranked list of local businesses as Google Search shows them. Use serpapi_google_maps_search for map-based search with zoom and rating filters.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `q` (`string`, required): What to look for, optionally with a place. Example: dentist near Mission District
- `async` (`boolean`, optional): Set true to submit the search and return at once with only search_metadata (id and status Processing); fetch the results later with serpapi_search_archive_get using search_metadata.id. Do not combine with no_cache. Example: false
- `device` (`string`, optional): Device type to emulate: desktop (default), tablet or mobile. Results and layout differ by device. One of: `desktop`, `tablet`, `mobile`.
- `gl` (`string`, optional): Two-letter country code (Google's gl parameter, lowercase) that sets which country's results to return. Example: us
- `google_domain` (`string`, optional): Google domain to search on. SerpApi uses google.com when omitted. Example: google.co.uk
- `hl` (`string`, optional): Language code for the results (Google's hl parameter): a two-letter language code, optionally with a region, such as en, fr, en-gb or es-419. Example: en
- `json_restrictor` (`string`, optional): Limit the response to the listed fields to keep it small, using SerpApi's JSON Restrictor syntax: comma-separated top-level keys, with dots, brackets and braces for nested fields. Example: organic_results[0:5].{title,link,snippet}
- `location` (`string`, optional): Place the search runs from, as a location name (city level works best, most specific first). Results are localized as if searched there. Find valid names with serpapi_locations_search. Example: Austin, Texas, United States
- `no_cache` (`boolean`, optional): Set true to force a fresh search instead of a cached result. SerpApi serves an identical search run in the last hour from its cache for free; a fresh search spends one search credit. Do not combine with async. Example: false
- `start` (`integer`, optional): Number of results to skip, for pagination: 0 is the first page, 20 the second, 40 the third. Each page is a separate search. SerpApi recommends following serpapi_pagination.next. Example: 20

Example input: `{"q":"<q>"}`

### `serpapi_google_maps_place_get`

Get Google Maps Place · Read-only

Get the Google Maps details of one place by its place_id or data_cid through SerpApi.
Returns place_results with title, place_id, data_id, data_cid, address, phone, website, gps_coordinates, rating, reviews, type, hours, price, popular times, user review highlights and images.
Use this when you already have a place id. Use serpapi_google_maps_search to find places and serpapi_google_maps_reviews_list for the full review list.
Get place_id or data_cid from serpapi_google_maps_search or serpapi_google_local_search.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `async` (`boolean`, optional): Set true to submit the search and return at once with only search_metadata (id and status Processing); fetch the results later with serpapi_search_archive_get using search_metadata.id. Do not combine with no_cache. Example: false
- `data_cid` (`string`, optional): Google customer ID (CID) of the place, from data_cid in serpapi_google_maps_search results. Set place_id or data_cid, not both. Example: 8756438412640839890
- `gl` (`string`, optional): Two-letter country code (Google's gl parameter, lowercase) that sets which country's results to return. Example: us
- `google_domain` (`string`, optional): Google domain to search on. SerpApi uses google.com when omitted. Example: google.co.uk
- `hl` (`string`, optional): Language code for the results (Google's hl parameter): a two-letter language code, optionally with a region, such as en, fr, en-gb or es-419. Example: en
- `json_restrictor` (`string`, optional): Limit the response to the listed fields to keep it small, using SerpApi's JSON Restrictor syntax: comma-separated top-level keys, with dots, brackets and braces for nested fields. Example: organic_results[0:5].{title,link,snippet}
- `no_cache` (`boolean`, optional): Set true to force a fresh search instead of a cached result. SerpApi serves an identical search run in the last hour from its cache for free; a fresh search spends one search credit. Do not combine with async. Example: false
- `place_id` (`string`, optional): Google place ID of the place, from a place_id in serpapi_google_maps_search results. Set place_id or data_cid, not both. Example: ChIJT2h1HKZZwokR0kgzEtsa03k

Example input: `{}`

### `serpapi_google_maps_reviews_list`

List Google Maps Reviews · Read-only

List the Google Maps reviews of one place through SerpApi, sorted and optionally filtered by topic or text.
Returns place_info (title, address, rating, reviews count), topics with topic_id, and reviews with user, rating, date, snippet, likes and owner response, plus serpapi_pagination.next_page_token.
Use this to read what customers say about a place. Use serpapi_google_maps_place_get for the place's details and review highlights.
Get data_id or place_id from serpapi_google_maps_search.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `async` (`boolean`, optional): Set true to submit the search and return at once with only search_metadata (id and status Processing); fetch the results later with serpapi_search_archive_get using search_metadata.id. Do not combine with no_cache. Example: false
- `data_id` (`string`, optional): Google Maps data ID of the place, from data_id in serpapi_google_maps_search results. Set data_id or place_id. Example: 0x89c259a61c75684f:0x79d31adb123348d2
- `hl` (`string`, optional): Language code for the results (Google's hl parameter): a two-letter language code, optionally with a region, such as en, fr, en-gb or es-419. Example: en
- `json_restrictor` (`string`, optional): Limit the response to the listed fields to keep it small, using SerpApi's JSON Restrictor syntax: comma-separated top-level keys, with dots, brackets and braces for nested fields. Example: organic_results[0:5].{title,link,snippet}
- `next_page_token` (`string`, optional): Token for the next page of reviews, from serpapi_pagination.next_page_token of the previous result. Example: CAESY0NBRVFDaHFBQVZIVWJLb1BqVm9VZFc4N1AxUlNiUQ
- `no_cache` (`boolean`, optional): Set true to force a fresh search instead of a cached result. SerpApi serves an identical search run in the last hour from its cache for free; a fresh search spends one search credit. Do not combine with async. Example: false
- `num` (`integer`, optional): Maximum number of reviews to return, 1 to 60 (default 10). Example: 20
- `place_id` (`string`, optional): Google place ID of the place, from place_id in serpapi_google_maps_search results. Set data_id or place_id. Example: ChIJT2h1HKZZwokR0kgzEtsa03k
- `query` (`string`, optional): Only reviews mentioning this text. Returns at most 5 reviews with no pagination. Cannot be combined with topic_id. Example: parking
- `sort_by` (`string`, optional): Review order: qualityScore most relevant (default), newestFirst, ratingHigh or ratingLow. One of: `qualityScore`, `newestFirst`, `ratingHigh`, `ratingLow`.
- `topic_id` (`string`, optional): Only reviews about one topic, using a topic_id from the topics list of an earlier result. Returns at most 5 reviews with no pagination. Cannot be combined with query. Example: /m/0k0m

Example input: `{}`

### `serpapi_google_maps_search`

Google Maps Search · Read-only

Search Google Maps through SerpApi for places matching a query near a location.
Returns local_results with title, place_id, data_id, data_cid, gps_coordinates, rating, reviews, price, type, address, phone, website, operating_hours and thumbnail, plus serpapi_pagination.
Use this to find businesses or places on a map. Use serpapi_google_maps_place_get for one place's full details, serpapi_google_maps_reviews_list for its reviews and serpapi_google_local_search for Google's local results list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `q` (`string`, required): What to look for on Google Maps, optionally with a place. Example: coffee shops in Brooklyn
- `async` (`boolean`, optional): Set true to submit the search and return at once with only search_metadata (id and status Processing); fetch the results later with serpapi_search_archive_get using search_metadata.id. Do not combine with no_cache. Example: false
- `google_domain` (`string`, optional): Google domain to search on. SerpApi uses google.com when omitted. Example: google.co.uk
- `hl` (`string`, optional): Language code for the results (Google's hl parameter): a two-letter language code, optionally with a region, such as en, fr, en-gb or es-419. Example: en
- `json_restrictor` (`string`, optional): Limit the response to the listed fields to keep it small, using SerpApi's JSON Restrictor syntax: comma-separated top-level keys, with dots, brackets and braces for nested fields. Example: organic_results[0:5].{title,link,snippet}
- `ll` (`string`, optional): Map center and zoom as @latitude,longitude,zoom (zoom 3z to 30z) or @latitude,longitude,height in meters. Cannot be combined with location. Example: @40.7455096,-74.0083012,14z
- `location` (`string`, optional): Place name to center the map on, used with z or m. Cannot be combined with ll. Example: Austin, Texas, United States
- `m` (`integer`, optional): Map height in meters, 1 to 15028132. Use instead of z with location. Example: 10000
- `min_rating` (`string`, optional): Prefer places rated at least this many stars: 2.0, 2.5, 3.0, 3.5, 4.0 or 4.5. Lower-rated places may still appear lower down. One of: `2.0`, `2.5`, `3.0`, `3.5`, `4.0`, `4.5`.
- `no_cache` (`boolean`, optional): Set true to force a fresh search instead of a cached result. SerpApi serves an identical search run in the last hour from its cache for free; a fresh search spends one search credit. Do not combine with async. Example: false
- `open_state` (`string`, optional): Only places open now (now) or open 24 hours (24h). One of: `now`, `24h`.
- `start` (`integer`, optional): Number of results to skip, for pagination: 0 is the first page, 20 the second, 40 the third. Each page is a separate search. SerpApi recommends staying at or below 100. Example: 20
- `z` (`integer`, optional): Map zoom level, 3 to 30 (effective maximum about 18 to 23). Required with location unless m is set. Example: 14

Example input: `{"q":"<q>"}`

### `serpapi_google_news_search`

Google News Search · Read-only

Search Google News through SerpApi, or open a news topic, publication or full-coverage story by its token.
Returns news_results with title, link, source (name, icon), date, thumbnail and nested stories, plus menu_links, related_topics and related_publications that carry topic_token and publication_token values.
Use this for recent news on a subject. Use serpapi_google_search for general web pages and serpapi_bing_news_search for Bing's news index.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `async` (`boolean`, optional): Set true to submit the search and return at once with only search_metadata (id and status Processing); fetch the results later with serpapi_search_archive_get using search_metadata.id. Do not combine with no_cache. Example: false
- `gl` (`string`, optional): Two-letter country code (Google's gl parameter, lowercase) that sets which country's results to return. SerpApi uses us when omitted. Example: us
- `hl` (`string`, optional): Language code for the results (Google's hl parameter): a two-letter language code, optionally with a region, such as en, fr, en-gb or es-419. Example: en
- `json_restrictor` (`string`, optional): Limit the response to the listed fields to keep it small, using SerpApi's JSON Restrictor syntax: comma-separated top-level keys, with dots, brackets and braces for nested fields. Example: organic_results[0:5].{title,link,snippet}
- `kgmid` (`string`, optional): Google Knowledge Graph ID of a topic or place (starts with /m/ or /g/) to get news about it. Must be used alone, without q or tokens. Example: /m/02mjmr
- `no_cache` (`boolean`, optional): Set true to force a fresh search instead of a cached result. SerpApi serves an identical search run in the last hour from its cache for free; a fresh search spends one search credit. Do not combine with async. Example: false
- `publication_token` (`string`, optional): Token of a publisher, taken from related_publications or a news result's source in an earlier result, to list that publisher's news. Can be combined with section_token. Example: CAAqBwgKMKHL9QowkqbaAg
- `q` (`string`, optional): The news search query. Leave empty when you pass topic_token, publication_token, story_token or kgmid instead; q cannot be combined with them. Example: semiconductor export rules
- `section_token` (`string`, optional): Token of a sub-section within a topic or publication, taken from an earlier result. Use with topic_token or publication_token. Example: CAQiS0NCQVNNZ29JTDIwdk1EZGpNWFlTQW1WdUdnSlZVeUlQQ0FRYUN3b0pMMjB2TURWeU5tY3pLZ0FxQ3dvSkVnZFNiMkp2ZEdsamN5Z0FQAVAB
- `so` (`integer`, optional): Sort order for a story_token or section_token result: 0 relevance (default) or 1 date. One of: `0`, `1`.
- `story_token` (`string`, optional): Token of a story, taken from a news result's story_token, to get the full coverage of that story from many outlets. Example: CAAqNggKIjBDQklTSGpvSmMzUnZjbmt0TXpZd1NoRUtEd2pBM2Y3YkJCRUFOZXo5WWI0QzhpZ0FQAQ
- `topic_token` (`string`, optional): Token of a Google News topic (such as Technology or World), taken from menu_links or related_topics in an earlier result. Can be combined with section_token. Example: CAAqJggKIiBDQkFTRWdvSUwyMHZNRGRqTVhZU0FtVnVHZ0pWVXlnQVAB

Example input: `{}`

### `serpapi_google_scholar_search`

Google Scholar Search · Read-only

Search Google Scholar through SerpApi for papers, or list the papers citing one article or all versions of it.
Returns organic_results with position, title, result_id, link, snippet, publication_info (authors, journal, year), resources (PDF links) and inline_links (cited_by total and cites_id, versions cluster_id), plus serpapi_pagination.
Use this for academic literature. Use serpapi_google_search for general web pages.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `as_rr` (`integer`, optional): Set 1 to show only review articles, 0 for all results (default). One of: `0`, `1`.
- `as_sdt` (`string`, optional): Search type and filter: 0 exclude patents (default), 7 include patents, 4 case law; for case law add comma-separated court codes after 4. Example: 4,33,192
- `as_vis` (`integer`, optional): Set 1 to exclude citation-only results, 0 to include them (default). One of: `0`, `1`.
- `as_yhi` (`integer`, optional): Latest publication year to include. Example: 2025
- `as_ylo` (`integer`, optional): Earliest publication year to include. Example: 2020
- `async` (`boolean`, optional): Set true to submit the search and return at once with only search_metadata (id and status Processing); fetch the results later with serpapi_search_archive_get using search_metadata.id. Do not combine with no_cache. Example: false
- `cites` (`string`, optional): Article ID whose citing papers to list (Cited By), from inline_links.cited_by.cites_id of an earlier result. With q, searches within the citing papers. Example: 1275980731835430123
- `cluster` (`string`, optional): Article ID whose versions to list (All Versions), from inline_links.versions.cluster_id of an earlier result. Cannot be combined with q or cites. Example: 1275980731835430123
- `hl` (`string`, optional): Language code for the results (Google's hl parameter): a two-letter language code, optionally with a region, such as en, fr, en-gb or es-419. Example: en
- `json_restrictor` (`string`, optional): Limit the response to the listed fields to keep it small, using SerpApi's JSON Restrictor syntax: comma-separated top-level keys, with dots, brackets and braces for nested fields. Example: organic_results[0:5].{title,link,snippet}
- `lr` (`string`, optional): Limit results to one or more languages, as lang_xx codes joined by |. Example: lang_en|lang_de
- `no_cache` (`boolean`, optional): Set true to force a fresh search instead of a cached result. SerpApi serves an identical search run in the last hour from its cache for free; a fresh search spends one search credit. Do not combine with async. Example: false
- `num` (`integer`, optional): Number of results to return, 1 to 20 (default 10). Example: 20
- `q` (`string`, optional): The Scholar query. Helpers such as author: and source: work. Required unless cites or cluster is set; not allowed with cluster. Example: transformer attention mechanism
- `scisbd` (`integer`, optional): Sort by date, showing only articles added in the past year: 0 sort by relevance (default), 1 by date with abstracts only, 2 by date with everything. One of: `0`, `1`, `2`.
- `start` (`integer`, optional): Number of results to skip, for pagination: 0 is the first page, 10 the second, 20 the third. Each page is a separate search. Example: 10

Example input: `{}`

### `serpapi_google_search`

Google Search · Read-only

Search Google web results through SerpApi.
Returns organic_results (position, title, link, snippet, source), plus answer_box, knowledge_graph, related_questions, local_results, top_stories, related_searches and serpapi_pagination when Google shows them, with search_metadata.id for the archive.
Use this for general web search. Use serpapi_google_news_search for news articles, serpapi_google_images_search for images and serpapi_google_shopping_search for products.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `q` (`string`, required): The search query, as you would type it into Google. Operators such as site:, intitle:, inurl: and quotes work. Example: best espresso machine 2026
- `async` (`boolean`, optional): Set true to submit the search and return at once with only search_metadata (id and status Processing); fetch the results later with serpapi_search_archive_get using search_metadata.id. Do not combine with no_cache. Example: false
- `device` (`string`, optional): Device type to emulate: desktop (default), tablet or mobile. Results and layout differ by device. One of: `desktop`, `tablet`, `mobile`.
- `gl` (`string`, optional): Two-letter country code (Google's gl parameter, lowercase) that sets which country's results to return. Example: us
- `google_domain` (`string`, optional): Google domain to search on. SerpApi uses google.com when omitted. Example: google.co.uk
- `hl` (`string`, optional): Language code for the results (Google's hl parameter): a two-letter language code, optionally with a region, such as en, fr, en-gb or es-419. Example: en
- `json_restrictor` (`string`, optional): Limit the response to the listed fields to keep it small, using SerpApi's JSON Restrictor syntax: comma-separated top-level keys, with dots, brackets and braces for nested fields. Example: organic_results[0:5].{title,link,snippet}
- `location` (`string`, optional): Place the search runs from, as a location name (city level works best, most specific first). Results are localized as if searched there. Find valid names with serpapi_locations_search. Example: Austin, Texas, United States
- `lr` (`string`, optional): Limit results to one or more languages, as lang_xx codes joined by |. Example: lang_en|lang_fr
- `no_cache` (`boolean`, optional): Set true to force a fresh search instead of a cached result. SerpApi serves an identical search run in the last hour from its cache for free; a fresh search spends one search credit. Do not combine with async. Example: false
- `safe` (`string`, optional): Adult content filter: active to filter explicit results, off to turn filtering off. Google blurs explicit content when omitted. One of: `active`, `off`.
- `start` (`integer`, optional): Number of results to skip, for pagination: 0 is the first page, 10 the second, 20 the third. Each page is a separate search. Example: 10
- `tbs` (`string`, optional): Advanced Google filter string (the tbs parameter), most often a time range: qdr:h past hour, qdr:d past 24 hours, qdr:w past week, qdr:m past month, qdr:y past year. Example: qdr:w

Example input: `{"q":"<q>"}`

### `serpapi_google_shopping_search`

Google Shopping Search · Read-only

Search Google Shopping through SerpApi for products and offers.
Returns shopping_results with position, title, product_id, product_link, source (store), price, extracted_price, old_price, rating, reviews, delivery and thumbnail, plus filters that carry shoprs tokens.
Use this to compare product prices across stores. Use serpapi_amazon_search for Amazon listings only.
Each call returns the first page only (about 40 products); SerpApi currently ignores pagination for this engine.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `async` (`boolean`, optional): Set true to submit the search and return at once with only search_metadata (id and status Processing); fetch the results later with serpapi_search_archive_get using search_metadata.id. Do not combine with no_cache. Example: false
- `device` (`string`, optional): Device type to emulate: desktop (default), tablet or mobile. Results and layout differ by device. One of: `desktop`, `tablet`, `mobile`.
- `free_shipping` (`boolean`, optional): Set true to show only products with free shipping. Example: true
- `gl` (`string`, optional): Two-letter country code (Google's gl parameter, lowercase) that sets which country's results to return. Example: us
- `google_domain` (`string`, optional): Google domain to search on. SerpApi uses google.com when omitted. Example: google.co.uk
- `hl` (`string`, optional): Language code for the results (Google's hl parameter): a two-letter language code, optionally with a region, such as en, fr, en-gb or es-419. Example: en
- `json_restrictor` (`string`, optional): Limit the response to the listed fields to keep it small, using SerpApi's JSON Restrictor syntax: comma-separated top-level keys, with dots, brackets and braces for nested fields. Example: organic_results[0:5].{title,link,snippet}
- `location` (`string`, optional): Place the search runs from, as a location name (city level works best, most specific first). Results are localized as if searched there. Find valid names with serpapi_locations_search. Example: Austin, Texas, United States
- `max_price` (`number`, optional): Highest price to include, in the local currency. Example: 300
- `min_price` (`number`, optional): Lowest price to include, in the local currency. Example: 50
- `no_cache` (`boolean`, optional): Set true to force a fresh search instead of a cached result. SerpApi serves an identical search run in the last hour from its cache for free; a fresh search spends one search credit. Do not combine with async. Example: false
- `on_sale` (`boolean`, optional): Set true to show only products on sale. Example: true
- `q` (`string`, optional): The product search query. Optional only when shoprs is set. Example: noise cancelling headphones
- `safe` (`string`, optional): Adult content filter: active to filter explicit results, off to turn filtering off. Google blurs explicit content when omitted. One of: `active`, `off`.
- `shoprs` (`string`, optional): Filter token from a filters entry of an earlier result; it carries the query and the chosen filters. Join several with ||. Example: CAEYASoTbm9pc2UgY2FuY2VsbGluZyBoZWFkcGhvbmVz
- `small_business` (`boolean`, optional): Set true to show only products from small businesses. Example: true
- `sort_by` (`integer`, optional): Sort order: 1 price low to high, 2 price high to low. Omit for relevance. One of: `1`, `2`.

Example input: `{}`

### `serpapi_google_trends_search`

Google Trends Search · Read-only

Get Google Trends data for one or more search terms through SerpApi: interest over time, interest by region, or related topics and queries.
Returns interest_over_time.timeline_data (date, values per query) for TIMESERIES, interest_by_region or compared_breakdown_by_region for GEO_MAP_0 and GEO_MAP, and related_topics or related_queries (rising and top) for those data types.
Use this to compare popularity or see what people search alongside a term. Use serpapi_google_autocomplete for live query suggestions.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `q` (`string`, required): Search term or terms, comma-separated (up to 5 for TIMESERIES and GEO_MAP, one for the other data types). Each term is at most 100 characters. Example: coffee,tea
- `async` (`boolean`, optional): Set true to submit the search and return at once with only search_metadata (id and status Processing); fetch the results later with serpapi_search_archive_get using search_metadata.id. Do not combine with no_cache. Example: false
- `cat` (`integer`, optional): Category ID to limit the data to, from Google Trends' category list. 0 (default) means all categories. Example: 71 (Food & Drink)
- `data_type` (`string`, optional): Kind of data: TIMESERIES interest over time (default), GEO_MAP comparison by region (2 to 5 terms), GEO_MAP_0 interest by region (one term), RELATED_TOPICS or RELATED_QUERIES (one term). One of: `TIMESERIES`, `GEO_MAP`, `GEO_MAP_0`, `RELATED_TOPICS`, `RELATED_QUERIES`.
- `date` (`string`, optional): Time range: now 1-H, now 4-H, now 1-d, now 7-d, today 1-m, today 3-m, today 12-m (default), today 5-y, all (since 2004), or a custom range yyyy-mm-dd yyyy-mm-dd. Example: today 3-m
- `geo` (`string`, optional): Location code, such as a country (US) or region (US-CA). Worldwide when omitted. Example: US
- `gprop` (`string`, optional): Google property to measure: web search when omitted, or images, news, froogle (Google Shopping) or youtube. One of: `images`, `news`, `froogle`, `youtube`.
- `hl` (`string`, optional): Language code for the results (Google's hl parameter): a two-letter language code, optionally with a region, such as en, fr, en-gb or es-419. Example: en
- `include_low_search_volume` (`boolean`, optional): Set true to include regions with low search volume. Only applies to GEO_MAP and GEO_MAP_0. Example: true
- `json_restrictor` (`string`, optional): Limit the response to the listed fields to keep it small, using SerpApi's JSON Restrictor syntax: comma-separated top-level keys, with dots, brackets and braces for nested fields. Example: organic_results[0:5].{title,link,snippet}
- `no_cache` (`boolean`, optional): Set true to force a fresh search instead of a cached result. SerpApi serves an identical search run in the last hour from its cache for free; a fresh search spends one search credit. Do not combine with async. Example: false
- `region` (`string`, optional): Region level for GEO_MAP and GEO_MAP_0: COUNTRY, REGION, DMA or CITY. Not every level has data for every location. One of: `COUNTRY`, `REGION`, `DMA`, `CITY`.
- `tz` (`integer`, optional): Time zone offset in minutes from UTC, with the sign reversed as in JavaScript (420 is UTC-7, -540 is UTC+9). Range -1439 to 1439; SerpApi uses 420 when omitted. Example: 0

Example input: `{"q":"<q>"}`

### `serpapi_locations_search`

Search Locations · Read-only

Find SerpApi location names that match a text, for the location input of the search tools. This call is free.
Returns a list ordered by reach, each with id, name, canonical_name, country_code, target_type (City, Region, Country and others), reach and gps.
Use this to get an exact canonical_name before setting location on serpapi_google_search, serpapi_google_local_search or other search tools.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `limit` (`integer`, optional): Maximum number of locations to return, up to 10. Example: 5
- `q` (`string`, optional): Text the location name must contain. Example: Austin

Example input: `{}`

### `serpapi_search_archive_get`

Get Archived Search · Read-only

Get the results of an earlier SerpApi search by its search ID, for up to 31 days after it ran.
Returns the same JSON the original search returned, with search_metadata.status of Queued, Processing, Success or Error; results are present once the status is Success.
Use this to collect the results of a search sent with async true, or to reread an earlier search without running it again.
Get the search ID from search_metadata.id of any SerpApi search tool's result.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `search_id` (`string`, required): ID of the earlier search, from search_metadata.id in its result. Example: 5b50d58a304bda2fca30bac9

Example input: `{"search_id":"<search_id>"}`

### `serpapi_youtube_search`

YouTube Search · Read-only

Search YouTube videos, channels and playlists through SerpApi.
Returns video_results with position_on_page, title, link, channel (name, link), published_date, views, length, description and thumbnail, plus channel_results, playlist_results, shorts_results and serpapi_pagination with an sp token for the next page.
Use this to find videos on a topic. Use serpapi_google_search with a video query for videos hosted elsewhere.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `search_query` (`string`, required): The YouTube search query. Example: how to repot a monstera
- `async` (`boolean`, optional): Set true to submit the search and return at once with only search_metadata (id and status Processing); fetch the results later with serpapi_search_archive_get using search_metadata.id. Do not combine with no_cache. Example: false
- `gl` (`string`, optional): Two-letter country code (Google's gl parameter, lowercase) that sets which country's results to return. Example: us
- `hl` (`string`, optional): Language code for the results (Google's hl parameter): a two-letter language code, optionally with a region, such as en, fr, en-gb or es-419. Example: en
- `json_restrictor` (`string`, optional): Limit the response to the listed fields to keep it small, using SerpApi's JSON Restrictor syntax: comma-separated top-level keys, with dots, brackets and braces for nested fields. Example: organic_results[0:5].{title,link,snippet}
- `no_cache` (`boolean`, optional): Set true to force a fresh search instead of a cached result. SerpApi serves an identical search run in the last hour from its cache for free; a fresh search spends one search credit. Do not combine with async. Example: false
- `sp` (`string`, optional): YouTube filter or pagination token: the sp value from serpapi_pagination of a previous result for the next page, or a filter copied from a YouTube URL's sp parameter, URL-decoded (CAI= sort by upload date, EgJwAQ== 4K, QgIIAQ== exact spelling). Pass it decoded, for example CAI= rather than CAI%3D, because the value is URL-encoded when sent. Example: CAI=

Example input: `{"search_query":"<search_query>"}`

## Related

Other Search connectors ([all 89](/agentkit/connectors/?category=search)):

- [Supadata](/agentkit/connectors/supadata/): Scalekit connector, API key, 21 tools
- [Exa](/agentkit/connectors/exa/): Scalekit connector, API key, 21 tools
- [Brave Search](/agentkit/connectors/brave/): Scalekit connector, API key, 18 tools


---

## More Scalekit documentation

| Resource | What it contains | When to use it |
|----------|-----------------|----------------|
| [/llms.txt](/llms.txt) | Structured index with routing hints per product area | Start here — find which documentation set covers your topic before loading full content |
| [/llms-full.txt](/llms-full.txt) | Complete documentation for all Scalekit products in one file | Use when you need exhaustive context across multiple products or when the topic spans several areas |
| [sitemap-0.xml](https://docs.scalekit.com/sitemap-0.xml) | Full URL list of every documentation page | Use to discover specific page URLs you can fetch for targeted, page-level answers |
