Skip to content
Scalekit Docs
Talk to an Engineer Dashboard

Connect AI agents to Agenty

Scalekit connector
Open markdown

The Agenty connector lets your AI agent act in each user's Agenty account. Each user connects their own Agenty access token once, and Scalekit sends it with every call, so your agent never handles credentials. It comes with 57 tools.

Tools
57
What they doRead · write · destructive
21 · 24 · 1221 read24 write12 destructive
Users sign in with

Setup

  1. Install the SDK

    Terminal window
    npm install @scalekit-sdk/node dotenv
  2. Set your credentials

    Add your Scalekit credentials to your .env file. Find values in app.scalekit.com > Developers > API Credentials.

    .env
    SCALEKIT_ENVIRONMENT_URL=<your-environment-url>
    SCALEKIT_CLIENT_ID=<your-client-id>
    SCALEKIT_CLIENT_SECRET=<your-client-secret>
  3. Create the Agenty connection

    In AgentKit > Connections, create an Agenty connection. The name you give it is the connection_name your code passes. See Configure connections.

  4. Authorize a user and make your first call

    quickstart.mts
    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 = 'agenty'
    const identifier = 'user_123'
    // Generate an authorization link for the user
    const { link } = await actions.getAuthorizationLink({ connectionName: connector, identifier })
    console.log('Authorize Agenty:', 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: 'agenty_agents_list',
    toolInput: {},
    })
    console.log(result)
    Terminal window
    npx tsx quickstart.mts

    Each user opens the link once and enters their Agenty credentials there. See Authorize a user for the full flow and statuses.

Tools

Pass the exact name to execute_tool
Try in PlaygroundRequest a tool
  • agenty_agent_getGet one agent's full configuration.Read-only

    Get Agent

    Get one agent's full configuration. Returns name, type, description, tags, start URL, collections and fields with their selectors, browser and proxy settings, and scripts. Use this before changing an agent with agenty_agent_update.

    Inputs

    agent_idstringrequired
    ID of the agent. Find it with agenty_agents_list. Example: abcd123456.
  • agenty_agent_input_getGet the input of an agent: the URLs or data it processes, or the list or agent it reads them from.Read-only

    Get Agent Input

    Get the input of an agent: the URLs or data it processes, or the list or agent it reads them from. Returns the input type and data. Use this before changing the input.

    Inputs

    agent_idstringrequired
    ID of the agent. Find it with agenty_agents_list. Example: abcd123456.
  • agenty_agent_result_getGet the latest scraped data of an agent as JSON, with paging, sorting and search, optionally for one collection.Read-only

    Get Agent Latest Result

    Get the latest scraped data of an agent as JSON, with paging, sorting and search, optionally for one collection. Returns the rows with their fields and the total count. Use this to read an agent's most recent data without looking up the job.

    Inputs

    agent_idstringrequired
    ID of the agent. Find it with agenty_agents_list. Example: abcd123456.
    collectioninteger
    Index of the collection to return when the agent has several (0 is the first). Example: 0.
    limitinteger
    Maximum number of rows to return. Example: 25.
    offsetinteger
    Number of rows to skip, for paging. Example: 0.
    orderstring
    Sort order: asc or desc. Example: desc.one of ascdesc
    searchstring
    Text to search for. Example: product.
    sortstring
    Field to sort by, e.g. created_at. Example: created_at.
  • agenty_agent_schedule_getGet the schedule of an agent.Read-only

    Get Agent Schedule

    Get the schedule of an agent. Returns the cron expression, description and whether it is enabled. Use this before changing the schedule.

    Inputs

    agent_idstringrequired
    ID of the agent. Find it with agenty_agents_list. Example: abcd123456.
  • agenty_agents_listList the web scraping, crawling and change detection agents in your Agenty account, with paging, sorting and search.Read-only

    List Agents

    List the web scraping, crawling and change detection agents in your Agenty account, with paging, sorting and search. Returns per agent its ID, name, type, description, tags, project and dates. Use this to find an agent. Use agenty_agent_get for one agent's full configuration.

    Inputs

    limitinteger
    Maximum number of agents to return. Example: 25.
    offsetinteger
    Number of agents to skip, for paging. Example: 0.
    orderstring
    Sort order: asc or desc. Example: desc.one of ascdesc
    searchstring
    Text to search for. Example: product.
    sortstring
    Field to sort by, e.g. created_at. Example: created_at.
  • agenty_connections_listList the third-party connections (Google Sheets, Dropbox, S3, Shopify and so on) that workflows can send data to.Read-only

    List Connections

    List the third-party connections (Google Sheets, Dropbox, S3, Shopify and so on) that workflows can send data to. Returns per connection its ID, name, type and status.

    Inputs

    This tool takes no inputs.

  • agenty_job_getGet the status and details of one job.Read-only

    Get Job

    Get the status and details of one job. Returns status (queued, started, completed, stopped, failed), pages processed and failed, credits used and timings. Use this to follow a job from agenty_job_start.

    Inputs

    job_idintegerrequired
    ID of the job (one agent run). Find it with agenty_jobs_list. Example: 1234567.
  • agenty_job_logs_listList the execution log lines of a job, with paging and search.Read-only

    List Job Logs

    List the execution log lines of a job, with paging and search. Returns timestamps, levels and messages. Use this to debug a failed or partial job.

    Inputs

    job_idintegerrequired
    ID of the job (one agent run). Find it with agenty_jobs_list. Example: 1234567.
    limitinteger
    Maximum number of log lines to return. Example: 25.
    offsetinteger
    Number of log lines to skip, for paging. Example: 0.
    orderstring
    Sort order: asc or desc. Example: desc.one of ascdesc
    searchstring
    Text to search for. Example: product.
    sortstring
    Field to sort by, e.g. created_at. Example: created_at.