> **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 Linkup

The Linkup connector lets your AI agent act in each user's Linkup account. Each user connects their own Linkup access token once, and Scalekit sends it with every call, so your agent never handles credentials. It comes with 11 tools.

**Authentication:** Bearer Token
**Categories:** Search, AI
**Tools:** 11: 9 read, 2 write, 0 destructive
**Users sign in with:** Access token
**Built by:** Scalekit connector
**Try it:** [Playground in the Scalekit dashboard](https://app.scalekit.com/ws/signup?sk_intent=playground&provider=LINKUP)

## 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 Linkup connection

   In **AgentKit > Connections**, create a Linkup 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 = 'linkup'
   const identifier = 'user_123'

   // Generate an authorization link for the user
   const { link } = await actions.getAuthorizationLink({ connectionName: connector, identifier })
   console.log('Authorize Linkup:', 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: 'linkup_credits_balance_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 = "linkup"
   identifier = "user_123"

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

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

   ```bash
   python quickstart.py
   ```

   Each user opens the link once and enters their Linkup 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: 'linkup_credits_balance_get',
  toolInput: {},
  connector: 'linkup',
  identifier: 'user_123',
})
```

**Python**

```python
result = actions.execute_tool(
    tool_name="linkup_credits_balance_get",
    tool_input={},
    connection_name="linkup",
    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.

### `linkup_credits_balance_get`

Get Credits Balance · Read-only

Get the remaining credit balance of your Linkup account.
Returns the balance.
Use this before large batches or deep searches to check you have credits.

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

Inputs: none.

Example input: `{}`

### `linkup_page_fetch`

Fetch Web Page · Read-only

Fetch one public web page or PDF by URL and return its content as clean markdown, optionally with typed JSON extracted by a schema.
Returns markdown, and when requested raw content, images or the structured object.
Use this when you already know the URL. Use linkup_search to find pages first.

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

Inputs:

- `url` (`string`, required): Full URL of the public page or PDF to fetch. Example: https://www.linkup.so/blog.
- `extract_images` (`boolean`, optional): Also return the images found on the page. Linkup defaults to false. Example: false.
- `include_raw_content` (`boolean`, optional): Also return the raw page content and its content type. Linkup defaults to false. Example: false.
- `instructions` (`string`, optional): Instructions that guide the structured extraction. Requires schema. Example: Extract the article title and publication date.
- `mode` (`string`, optional): Fetch strategy. pro has much higher success on hard-to-retrieve pages and costs more. Linkup defaults to standard. Example: standard. One of: `standard`, `pro`.
- `render_js` (`boolean`, optional): Render the page's JavaScript before extracting. Slower; use for single-page apps. Linkup defaults to false. Example: false.
- `schema` (`string`, optional): JSON Schema, as a string, of type object describing structured data to extract from the page. Fields with no grounded value are left out. Example: {"type":"object","properties":{"title":{"type":"string"}}}

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

### `linkup_research_get`

Get Research Task · Read-only

Get one research task by id, with its status and, once completed, its answer or structured output and sources.
Returns the task id, status (pending, processing, completed or failed) and output.
Use this to poll a task started by linkup_research_create. Use linkup_research_list to find task ids.

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

Inputs:

- `id` (`string`, required): Research task id returned by linkup_research_create or linkup_research_list. Example: 6f1c2a9e-3b4d-4e5f-8a7b-1c2d3e4f5a6b.

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

### `linkup_research_list`

List Research Tasks · Read-only

List the research tasks created by your Linkup organization, with pagination.
Returns tasks with their id, status and timestamps.
Use this to find a research task id. Use linkup_research_get for one task's output.

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

Inputs:

- `page` (`integer`, optional): Page number, starting at 1. Linkup defaults to 1. Example: 1.
- `page_size` (`integer`, optional): Results per page, 1 to 100. Linkup defaults to 10. Example: 20.
- `sort_by` (`string`, optional): Field to sort by. Example: createdAt. One of: `createdAt`, `updatedAt`.
- `sort_direction` (`string`, optional): Sort direction. Linkup defaults to asc. Example: desc. One of: `asc`, `desc`.

Example input: `{}`

### `linkup_rewind_fetch`

Fetch Past Page (Rewind) · Read-only

Return a web page as it was at the end of a given UTC day (beta).
Returns the stored page content for that day.
Use this to see a page's past version. The URL must match the stored URL exactly. Use linkup_page_fetch for the live page.

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

Inputs:

- `as_of` (`string`, required): UTC day to fetch the page as of, YYYY-MM-DD. Example: 2025-09-17.
- `url` (`string`, required): Exact URL of the page. No normalization is applied, so it must match the stored URL. Example: https://www.federalreserve.gov/newsevents/pressreleases.htm.

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

### `linkup_rewind_search`

Search Past Web (Rewind) · Read-only

Search the web as it was at the end of a given UTC day (beta).
Returns ranked results from Linkup's stored snapshot for that day.
Use this for point-in-time research or backtesting. Use linkup_search for the live web.

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

Inputs:

- `as_of` (`string`, required): UTC day to search as of, YYYY-MM-DD. Example: 2025-09-17.
- `q` (`string`, required): The search query. Example: Federal Reserve interest rate decision.

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

### `linkup_search`

Search the Web · Read-only

Search the live web with Linkup and get ranked sources, a sourced answer, or JSON shaped by your schema.
Returns results (name, url, content snippets) for searchResults, an answer with sources for sourcedAnswer, or an object matching structured_output_schema.
Use this for current facts and source discovery. Use linkup_page_fetch to read one known URL in full, and linkup_research_create for multi-step research that takes minutes.

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

Inputs:

- `depth` (`string`, required): Search depth. flash is lowest latency (sources and snippets in a few hundred ms), fast is quick single-pass, standard is the default balance, deep iterates with more sources and costs more credits. Example: standard. One of: `flash`, `fast`, `standard`, `deep`.
- `output_type` (`string`, required): Response shape: searchResults (ranked sources with snippets), sourcedAnswer (natural-language answer with sources) or structured (JSON matching structured_output_schema). Example: sourcedAnswer. One of: `searchResults`, `sourcedAnswer`, `structured`.
- `q` (`string`, required): Natural-language question or query to search for. Be specific about entities, dates and the facts wanted. Example: What was Nvidia's revenue in Q2 fiscal 2026?.
- `exclude_domains` (`array`, optional): Never return results from these domains. Example: ["wikipedia.org"]
- `from_date` (`string`, optional): Only consider results published on or after this date (YYYY-MM-DD). Must be before to_date. Example: 2026-01-01.
- `include_domains` (`array`, optional): Only search these domains (up to 100). Omit to search the whole web. Example: ["sec.gov", "reuters.com"]
- `include_images` (`boolean`, optional): Include image results. Linkup defaults to false. Example: false.
- `include_inline_citations` (`boolean`, optional): For sourcedAnswer only: add inline citations to the answer text. Linkup defaults to false. Example: true.
- `include_sources` (`boolean`, optional): For structured output only: also return the sources used. This changes the response shape. Linkup defaults to false. Example: true.
- `max_results` (`integer`, optional): Maximum number of results to return. Example: 10.
- `structured_output_schema` (`string`, optional): JSON Schema, as a string, for the response when output_type is structured. Required only for structured output; the root must be of type object. Example: {"type":"object","properties":{"revenue":{"type":"string"}}}
- `to_date` (`string`, optional): Only consider results published on or before this date (YYYY-MM-DD). Must be after from_date. Example: 2026-06-30.

Example input: `{"depth":"flash","output_type":"searchResults","q":"<q>"}`

### `linkup_task_get`

Get Batch Task · Read-only

Get one batch task by id, with its type, status and, once completed, its output.
Returns the task id, type, status and output.
Use this to poll a task created by linkup_tasks_create. Use linkup_tasks_list to find ids.

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

Inputs:

- `id` (`string`, required): Task id returned by linkup_tasks_create or linkup_tasks_list. Example: 6f1c2a9e-3b4d-4e5f-8a7b-1c2d3e4f5a6b.

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

### `linkup_tasks_list`

List Batch Tasks · Read-only

List batch tasks for your Linkup organization, optionally filtered by type and status, with pagination.
Returns tasks with id, type, status and timestamps.
Use this to check progress across a batch. Use linkup_task_get for one task's output.

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

Inputs:

- `page` (`integer`, optional): Page number, starting at 1. Linkup defaults to 1. Example: 1.
- `page_size` (`integer`, optional): Results per page, 1 to 100. Linkup defaults to 10. Example: 20.
- `sort_by` (`string`, optional): Field to sort by. Example: createdAt. One of: `createdAt`, `updatedAt`.
- `sort_direction` (`string`, optional): Sort direction. Linkup defaults to asc. Example: desc. One of: `asc`, `desc`.
- `status` (`array`, optional): Only tasks with these statuses: pending, processing, completed or failed. Example: ["completed"]
- `type` (`array`, optional): Only tasks of these types: search, fetch, research or extract. Example: ["search"]

Example input: `{}`

### `linkup_research_create`

Create Research Task · Write

Start an asynchronous Linkup research task: an agent browses the web over several steps and writes a sourced answer or structured result.
Returns the research task id and status; the result is not in this response.
Use this for questions that need many sources or a report. Poll linkup_research_get with the id until status is completed. Use linkup_search for quick lookups.

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

Inputs:

- `output_type` (`string`, required): sourcedAnswer for a written answer with sources, or structured for JSON matching structured_output_schema. Example: sourcedAnswer. One of: `sourcedAnswer`, `structured`.
- `q` (`string`, required): The research question, stated fully with the scope and the output you want. Example: Compare the pricing and API limits of the top five web search APIs for AI agents in 2026.
- `exclude_domains` (`array`, optional): Never return results from these domains. Example: ["wikipedia.org"]
- `from_date` (`string`, optional): Only consider results published on or after this date (YYYY-MM-DD). Must be before to_date. Example: 2026-01-01.
- `include_domains` (`array`, optional): Only search these domains (up to 100). Omit to search the whole web. Example: ["sec.gov", "reuters.com"]
- `mode` (`string`, optional): Pin the research agent mode: answer, investigate or research. Omit (auto) to let the agent pick from the question. Example: research. One of: `auto`, `answer`, `investigate`, `research`.
- `reasoning_depth` (`string`, optional): Reasoning depth, S < M < L < XL. Higher depths are slower and more thorough. Linkup defaults to L. Example: M. One of: `S`, `M`, `L`, `XL`.
- `structured_output_schema` (`string`, optional): JSON Schema, as a string, for the response when output_type is structured. Required only for structured output; the root must be of type object. Example: {"type":"object","properties":{"revenue":{"type":"string"}}}
- `to_date` (`string`, optional): Only consider results published on or before this date (YYYY-MM-DD). Must be after from_date. Example: 2026-06-30.

Example input: `{"output_type":"sourcedAnswer","q":"<q>"}`

### `linkup_tasks_create`

Create Batch Tasks · Write

Queue up to 100 search, fetch or research calls to run asynchronously in one batch.
Returns the created task ids; results are not in this response.
Use this for bulk work. Poll linkup_task_get per id or linkup_tasks_list. Use linkup_search or linkup_page_fetch for a single synchronous call.

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

Inputs:

- `tasks` (`array`, required): Array of 1 to 100 task objects. Each has a type (search, fetch or research) and an input object that takes the same fields as that endpoint's body, for example q, depth and outputType for search, or url for fetch. Example: [{"type":"search","input":{"q":"Linkup API pricing","depth":"standard","outputType":"searchResults"}}]

Example input: `{"tasks":[{"type":"search","input":{"q":"Linkup API pricing","depth":"standard","outputType":"searchResults"}}]}`

## 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 |
