> **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 OpenGraph.io

The OpenGraph.io connector lets your AI agent act in each user's OpenGraph.io account. Each user connects their own OpenGraph.io API key once, and Scalekit sends it with every call, so your agent never handles credentials. It comes with 8 tools.

**Authentication:** API Key
**Categories:** Search, Developer Tools
**Tools:** 8: 5 read, 3 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=OPENGRAPHIO)

## 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 OpenGraph.io connection

   In **AgentKit > Connections**, create an OpenGraph.io 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 = 'opengraphio'
   const identifier = 'user_123'

   // Generate an authorization link for the user
   const { link } = await actions.getAuthorizationLink({ connectionName: connector, identifier })
   console.log('Authorize OpenGraph.io:', 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: 'opengraphio_markdown_get',
     toolInput: { encoded_url: 'https%3A%2F%2Fgithub.com' },
   })
   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 = "opengraphio"
   identifier = "user_123"

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

   # Make your first call
   result = actions.execute_tool(
       tool_input={"encoded_url":"https%3A%2F%2Fgithub.com"},
       tool_name="opengraphio_markdown_get",
       connection_name=connection_name,
       identifier=identifier,
   )
   print(result)
   ```

   ```bash
   python quickstart.py
   ```

   Each user opens the link once and enters their OpenGraph.io 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: 'opengraphio_markdown_get',
  toolInput: {
    encoded_url: '<encoded_url>',
  },
  connector: 'opengraphio',
  identifier: 'user_123',
})
```

**Python**

```python
result = actions.execute_tool(
    tool_name="opengraphio_markdown_get",
    tool_input={
        "encoded_url": "<encoded_url>",
    },
    connection_name="opengraphio",
    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.

### `opengraphio_markdown_get`

Get Page as Markdown · Read-only

Convert a web page into clean Markdown, optionally chunked for embeddings or filtered to passages matching a question.
Returns the Markdown as text, or with format json an envelope with markdown, metadata, chunks, headings, links and images.
Use this to read a page's content. Use opengraphio_scrape_html for raw HTML or opengraphio_query for an AI answer.

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

Inputs:

- `encoded_url` (`string`, required): Target page URL, URL-encoded as one path segment (encode : as %3A and / as %2F, as JavaScript encodeURIComponent does). Send the encoded form, not the raw URL. Example: https%3A%2F%2Fgithub.com.
- `accept_lang` (`string`, optional): Accept-Language header sent to the page. Defaults to en-US. Example: en-US.
- `ai_sanitize` (`boolean`, optional): Scan the page for prompt-injection text (1 extra credit). Defaults to false. Example: false.
- `ai_sanitize_mode` (`string`, optional): What to do when ai_sanitize finds something: sanitize (strip it, the default), warn or block. Example: sanitize. One of: `sanitize`, `warn`, `block`.
- `auto_proxy` (`boolean`, optional): Let OpenGraph.io pick the best proxy for the target domain. Defaults to true. Example: true.
- `auto_render` (`boolean`, optional): Let OpenGraph.io turn on JavaScript rendering when the page needs it. Defaults to true. Example: true.
- `cache_ok` (`boolean`, optional): Allow a cached result. Defaults to true; set false to force a fresh fetch. Example: true.
- `chunk_overlap` (`integer`, optional): Characters repeated from the end of the previous chunk, at most half of chunk_size. Defaults to 0. Example: 200.
- `chunk_size` (`integer`, optional): Target chunk size in characters, 200 to 20000. Defaults to 2000. Example: 2000.
- `chunking` (`boolean`, optional): Split the document into chunks for a vector database. Needs format json. Defaults to false. Example: false.
- `exclude_tags` (`string`, optional): Comma-separated CSS selectors to drop. Example: nav,.sidebar,footer.
- `format` (`string`, optional): markdown (default, plain Markdown text) or json (an envelope with metadata and optional chunks). Use json for chunking, headings, links or images. Example: json. One of: `markdown`, `json`.
- `full_render` (`boolean`, optional): Always render the page in a headless browser before reading it, for single-page apps. Defaults to false; adds time and credits. Example: false.
- `include_headings` (`boolean`, optional): Return the heading outline and counts of tables and code blocks (format json). Defaults to false. Example: false.
- `include_images` (`boolean`, optional): Return every image with its source and alt text (format json). Defaults to false. Example: false.
- `include_links` (`boolean`, optional): Return every link in the main content as an absolute URL (format json). Defaults to false. Example: false.
- `include_tags` (`string`, optional): Comma-separated CSS selectors to keep. Example: article,main.
- `max_chars` (`integer`, optional): Maximum Markdown length in characters. Example: 20000.
- `only_main_content` (`boolean`, optional): Drop navigation, sidebars and footers. Defaults to true. Example: true.
- `proxy_country` (`string`, optional): Two-letter country code to fetch from when a proxy is used. Example: US.
- `query` (`string`, optional): Keywords or a question to keep only the matching passages (keyword matching, no AI cost), up to 512 characters. Example: pricing plans.
- `query_top_k` (`integer`, optional): Number of matching passages to return with query. Defaults to 5. Example: 5.
- `retry` (`boolean`, optional): Retry a failed fetch, escalating to stronger proxy tiers. Defaults to true; you are billed only for the tier that succeeds. Example: true.
- `use_premium` (`boolean`, optional): Fetch through a residential proxy (needs a plan that includes it). Defaults to false. Example: false.
- `use_proxy` (`boolean`, optional): Fetch through a standard datacenter proxy. Defaults to false. Example: false.
- `use_superior` (`boolean`, optional): Fetch through a mobile proxy, the highest success rate (needs a plan that includes it). Defaults to false. Example: false.

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

### `opengraphio_oembed_get`

Get oEmbed · Read-only

Get an oEmbed embed for any URL: the provider's native embed (YouTube, Vimeo and others) or a hosted OpenGraph.io preview card.
Returns oEmbed fields (type, html, width, height, title, provider_name, thumbnail_url) and source (native, discovery or og_frame).
Use this to embed a link in a page. Use opengraphio_site_get for the raw metadata.

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

Inputs:

- `url` (`string`, required): Full URL to embed (not encoded). Example: https://www.youtube.com/watch?v=dQw4w9WgXcQ.
- `always_og_frame` (`boolean`, optional): Always return a hosted OpenGraph.io card, even when a native embed exists. Defaults to false. Example: false.
- `auto_proxy` (`boolean`, optional): Let OpenGraph.io pick the best proxy for the target domain. Defaults to true. Example: true.
- `auto_render` (`boolean`, optional): Let OpenGraph.io turn on JavaScript rendering when the page needs it. Defaults to true. Example: true.
- `cache_ok` (`boolean`, optional): Allow a cached result. Defaults to true; set false to force a fresh fetch. Example: true.
- `full_render` (`boolean`, optional): Always render the page in a headless browser before reading it, for single-page apps. Defaults to false; adds time and credits. Example: false.
- `max_cache_age` (`integer`, optional): Maximum age of cached page metadata in milliseconds. Defaults to 432000000. Example: 432000.
- `maxheight` (`integer`, optional): Maximum embed height in pixels. Example: 480.
- `maxwidth` (`integer`, optional): Maximum embed width in pixels. Example: 640.
- `orientation` (`string`, optional): Fallback card orientation: vertical (800x600, the default) or horizontal (400x150). Example: vertical. One of: `vertical`, `horizontal`.
- `proxy_country` (`string`, optional): Two-letter country code to fetch from when a proxy is used. Example: US.
- `retry` (`boolean`, optional): Retry a failed fetch, escalating to stronger proxy tiers. Defaults to true; you are billed only for the tier that succeeds. Example: true.
- `template` (`string`, optional): Fallback card layout: default, compact or horizontal. Overrides orientation. Example: compact. One of: `default`, `compact`, `horizontal`.
- `theme` (`string`, optional): Fallback card colour scheme: light, dark or auto (the default). Example: auto. One of: `light`, `dark`, `auto`.
- `use_premium` (`boolean`, optional): Fetch through a residential proxy (needs a plan that includes it). Defaults to false. Example: false.
- `use_proxy` (`boolean`, optional): Fetch through a standard datacenter proxy. Defaults to false. Example: false.
- `use_superior` (`boolean`, optional): Fetch through a mobile proxy, the highest success rate (needs a plan that includes it). Defaults to false. Example: false.

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

### `opengraphio_scrape_html`

Scrape Page HTML · Read-only

Fetch the raw HTML of a web page, with optional JavaScript rendering and proxies.
Returns the page HTML as text.
Use this when you need the markup. Use opengraphio_markdown_get for clean text or opengraphio_extract for chosen elements.

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

Inputs:

- `encoded_url` (`string`, required): Target page URL, URL-encoded as one path segment (encode : as %3A and / as %2F, as JavaScript encodeURIComponent does). Send the encoded form, not the raw URL. Example: https%3A%2F%2Fgithub.com.
- `accept_lang` (`string`, optional): Accept-Language header sent to the page. Defaults to en-US. Example: en-US.
- `auto_proxy` (`boolean`, optional): Let OpenGraph.io pick the best proxy for the target domain. Defaults to true. Example: true.
- `auto_render` (`boolean`, optional): Let OpenGraph.io turn on JavaScript rendering when the page needs it. Defaults to true. Example: true.
- `cache_ok` (`boolean`, optional): Allow a cached result. Defaults to true; set false to force a fresh fetch. Example: true.
- `full_render` (`boolean`, optional): Always render the page in a headless browser before reading it, for single-page apps. Defaults to false; adds time and credits. Example: false.
- `retry` (`boolean`, optional): Retry a failed fetch, escalating to stronger proxy tiers. Defaults to true; you are billed only for the tier that succeeds. Example: true.
- `use_premium` (`boolean`, optional): Fetch through a residential proxy (needs a plan that includes it). Defaults to false. Example: false.
- `use_proxy` (`boolean`, optional): Fetch through a standard datacenter proxy. Defaults to false. Example: false.
- `use_superior` (`boolean`, optional): Fetch through a mobile proxy, the highest success rate (needs a plan that includes it). Defaults to false. Example: false.

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

### `opengraphio_screenshot_capture`

Capture Screenshot · Read-only

Capture a screenshot of a web page or one element on it.
Returns screenshotUrl (a temporary image URL that expires after 24 hours), dimensions and requestInfo.
Use this for a visual snapshot. Use opengraphio_site_get for the page's own preview image.

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

Inputs:

- `encoded_url` (`string`, required): Target page URL, URL-encoded as one path segment (encode : as %3A and / as %2F, as JavaScript encodeURIComponent does). Send the encoded form, not the raw URL. Example: https%3A%2F%2Fgithub.com.
- `block_cookie_banner` (`boolean`, optional): Hide known cookie consent banners. Defaults to true. Example: true.
- `cache_ok` (`boolean`, optional): Allow a cached screenshot. Defaults to true. Example: true.
- `capture_delay` (`integer`, optional): Milliseconds to wait before capturing, 0 to 10000. Example: 1000.
- `dark_mode` (`boolean`, optional): Set the browser to prefer dark mode. Defaults to false. Example: false.
- `dimensions` (`string`, optional): Viewport size: xs (375x812 phone), sm (1024x768 tablet), md (1366x768 laptop, the default) or lg (1920x1080 desktop). Example: lg. One of: `xs`, `sm`, `md`, `lg`.
- `exclude_selectors` (`string`, optional): Comma-separated CSS selectors to hide before capturing. Example: .header,.footer.
- `format` (`string`, optional): Image format. Defaults to jpeg. Example: png. One of: `jpeg`, `png`, `webp`.
- `full_page` (`boolean`, optional): Capture the whole scrollable page instead of only the viewport. Defaults to false. Example: false.
- `navigation_timeout` (`integer`, optional): Page navigation timeout in milliseconds, 1000 to 60000. Defaults to 30000. Example: 30000.
- `quality` (`integer`, optional): Image quality from 10 to 80, rounded to the nearest 10. Defaults to 80. Example: 80.
- `selector` (`string`, optional): CSS selector of a single element to capture. Example: .main-content.
- `use_proxy` (`boolean`, optional): Fetch through a proxy for protected sites. Defaults to false. Example: false.

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

### `opengraphio_site_get`

Get Site Metadata (Unfurl) · Read-only

Unfurl a URL: read its Open Graph tags, Twitter Card and HTML meta tags to build a link preview.
Returns hybridGraph (best merged title, description, image, url, site_name, type and favicon), openGraph, twitterCard, htmlInferred and requestInfo.
Use this for link-preview metadata. Use opengraphio_markdown_get for the page text or opengraphio_scrape_html for raw HTML.

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

Inputs:

- `encoded_url` (`string`, required): Target page URL, URL-encoded as one path segment (encode : as %3A and / as %2F, as JavaScript encodeURIComponent does). Send the encoded form, not the raw URL. Example: https%3A%2F%2Fgithub.com.
- `accept_lang` (`string`, optional): Accept-Language header sent to the page. Defaults to en-US. Example: en-US.
- `auto_proxy` (`boolean`, optional): Let OpenGraph.io pick the best proxy for the target domain. Defaults to true. Example: true.
- `auto_render` (`boolean`, optional): Let OpenGraph.io turn on JavaScript rendering when the page needs it. Defaults to true. Example: true.
- `cache_ok` (`boolean`, optional): Allow a cached result. Defaults to true; set false to force a fresh fetch. Example: true.
- `full_render` (`boolean`, optional): Always render the page in a headless browser before reading it, for single-page apps. Defaults to false; adds time and credits. Example: false.
- `max_cache_age` (`integer`, optional): Maximum cache age in seconds. Defaults to 432000 (5 days). Example: 432000.
- `proxy_country` (`string`, optional): Two-letter country code to fetch from when a proxy is used. Example: US.
- `retry` (`boolean`, optional): Retry a failed fetch, escalating to stronger proxy tiers. Defaults to true; you are billed only for the tier that succeeds. Example: true.
- `use_ai` (`boolean`, optional): Use AI to improve metadata extraction. Defaults to false. Example: false.
- `use_premium` (`boolean`, optional): Fetch through a residential proxy (needs a plan that includes it). Defaults to false. Example: false.
- `use_proxy` (`boolean`, optional): Fetch through a standard datacenter proxy. Defaults to false. Example: false.
- `use_superior` (`boolean`, optional): Fetch through a mobile proxy, the highest success rate (needs a plan that includes it). Defaults to false. Example: false.

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

### `opengraphio_brand_analyze`

Analyze Brand · Write

Analyze a company website and return its brand profile, reading several pages of the site.
Returns brand sections (such as identity, offering, audience, visual and links) and the answers to any custom questions, as JSON or Markdown.
Use this to profile a company from its site. A fresh analysis can take tens of seconds.

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

Inputs:

- `encoded_url` (`string`, required): Target page URL, URL-encoded as one path segment (encode : as %3A and / as %2F, as JavaScript encodeURIComponent does). Send the encoded form, not the raw URL. Example: https%3A%2F%2Fgithub.com.
- `cache_ok` (`boolean`, optional): Allow a cached analysis. Defaults to true; set false for a fresh run. Example: true.
- `format` (`string`, optional): json (default) or markdown. Example: json. One of: `json`, `markdown`.
- `include_meta` (`boolean`, optional): Add a _meta object with the source of each field. Defaults to false. Example: false.
- `questions` (`array`, optional): Up to 3 custom questions, each an object with a key, the question and a type such as string or string[]. Example: [{"key": "buying_triggers", "question": "What situations would make someone buy from this company?", "type": "string[]"}]
- `sections` (`object`, optional): Boolean flags keyed by section name; set a section to false to skip it. All sections are on by default. Example: {"visual": false, "links": false}

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

### `opengraphio_extract`

Extract Page Elements · Write

Scrape a page and extract the elements you choose with CSS selectors, plus the concatenated text of common tags.
Returns the extracted values keyed by your selector names, the tag items and concatenatedText.
Use this for structured fields from a page. Use opengraphio_query to ask a question in plain language instead.

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

Inputs:

- `site` (`string`, required): Full URL of the page to scrape (not encoded). Example: https://example.com/blog.
- `auto_proxy` (`boolean`, optional): Let OpenGraph.io pick the best proxy for the target domain. Defaults to true. Example: true.
- `auto_render` (`boolean`, optional): Let OpenGraph.io turn on JavaScript rendering when the page needs it. Defaults to true. Example: true.
- `cache_ok` (`boolean`, optional): Allow a cached result. Defaults to true; set false to force a fresh fetch. Example: true.
- `full_render` (`boolean`, optional): Always render the page in a headless browser before reading it, for single-page apps. Defaults to false; adds time and credits. Example: false.
- `html_elements` (`string`, optional): Comma-separated HTML tags used to build concatenatedText. Defaults to title,h1,h2,h3,h4,h5,p. Example: title,h1,p.
- `load_more_clicks` (`integer`, optional): How many times to click the load-more button, 1 to 10. Defaults to 3. Example: 3.
- `load_more_selector` (`string`, optional): CSS selector of a Load more button to click before capturing; turns on JavaScript rendering. Example: button.load-more.
- `max_cache_age` (`integer`, optional): Maximum cache age in seconds. Example: 432000.
- `retry` (`boolean`, optional): Retry a failed fetch, escalating to stronger proxy tiers. Defaults to true; you are billed only for the tier that succeeds. Example: true.
- `scroll_to_bottom` (`boolean`, optional): Scroll the whole page first to load lazy content; turns on JavaScript rendering. Defaults to false. Example: false.
- `selectors` (`object`, optional): Map of your own keys to CSS selector settings. Each setting has a selector, multiple (true returns every match as a list), type (text, or attr for an attribute) and attr (the attribute name when type is attr). Example: {"title": {"selector": "h1", "type": "text"}, "links": {"selector": "a", "multiple": true, "type": "attr", "attr": "href"}}.
- `use_premium` (`boolean`, optional): Fetch through a residential proxy (needs a plan that includes it). Defaults to false. Example: false.
- `use_proxy` (`boolean`, optional): Fetch through a standard datacenter proxy. Defaults to false. Example: false.
- `use_superior` (`boolean`, optional): Fetch through a mobile proxy, the highest success rate (needs a plan that includes it). Defaults to false. Example: false.
- `wait_for_selector` (`string`, optional): CSS selector to wait for before capturing; turns on JavaScript rendering. Example: #content-loaded.

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

### `opengraphio_query`

Ask a Question About a Page · Write

Ask an AI question about a web page and get the answer, optionally in a JSON shape you define.
Returns answer (text, or an object matching your response structure) and requestInfo.
Use this to pull an answer or summary from a page in plain language. Use opengraphio_extract for exact CSS-selected values. Each fresh answer costs 100 credits (nano) or 200 (mini).

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

Inputs:

- `encoded_url` (`string`, required): Target page URL, URL-encoded as one path segment (encode : as %3A and / as %2F, as JavaScript encodeURIComponent does). Send the encoded form, not the raw URL. Example: https%3A%2F%2Fgithub.com.
- `query` (`string`, required): Question to ask about the page. Example: What products does this company sell?.
- `auto_proxy` (`boolean`, optional): Let OpenGraph.io pick the best proxy for the target domain. Defaults to true. Example: true.
- `auto_render` (`boolean`, optional): Let OpenGraph.io turn on JavaScript rendering when the page needs it. Defaults to true. Example: true.
- `cache_ok` (`boolean`, optional): Allow a cached result. Defaults to true; set false to force a fresh fetch. Example: true.
- `full_render` (`boolean`, optional): Always render the page in a headless browser before reading it, for single-page apps. Defaults to false; adds time and credits. Example: false.
- `model_size` (`string`, optional): AI model size: nano (default, 100 credits) or mini (larger, 200 credits). Example: nano. One of: `nano`, `mini`.
- `response_structure` (`object`, optional): Optional JSON schema for a structured answer. Example: {"headline": "string", "products": ["string"]}
- `retry` (`boolean`, optional): Retry a failed fetch, escalating to stronger proxy tiers. Defaults to true; you are billed only for the tier that succeeds. Example: true.
- `use_premium` (`boolean`, optional): Fetch through a residential proxy (needs a plan that includes it). Defaults to false. Example: false.
- `use_proxy` (`boolean`, optional): Fetch through a standard datacenter proxy. Defaults to false. Example: false.
- `use_superior` (`boolean`, optional): Fetch through a mobile proxy, the highest success rate (needs a plan that includes it). Defaults to false. Example: false.

Example input: `{"encoded_url":"<encoded_url>","query":"<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 |
