Getting started

From an Actor to useful data.

Start in Apify Console, inspect a small result, and then connect the feed to your application.

  1. 01 / Choose

    Pick your sources.

    Use one board for a focused feed, or a bundle for a broader search.

  2. 02 / Try

    Start with a few jobs.

    Open the Actor on Apify, check its input and pricing, and set a small result limit and a cost cap.

  3. 03 / Connect

    Make the data useful.

    Export the results, map them into your app, or connect a spreadsheet. Test the destination before scheduling.

Choose an Actor →

For your application

Use the API with a bounded run.

Read the selected Actor’s current input schema. Replace the example Actor and input.json with your selected Actor’s identity and validated input.

REST / start a run
curl --request POST \
  'https://api.apify.com/v2/acts/nomad-agent~all-jobs-scraper/runs?build=latest&timeout=300&maxTotalChargeUsd=0.20' \
  --header "Authorization: Bearer $APIFY_TOKEN" \
  --header 'Content-Type: application/json' \
  --data-binary @input.json

The $0.20 cap and 300-second timeout are example guardrails, not a guarantee that every source scan completes. Read current pricing before choosing your budget. Keep credentials in your environment, never in URLs or shared files.

  1. Retain the returned run ID and poll it to a terminal status with a finite deadline.
  2. Record the immutable buildId returned by the run. Read the default dataset and RUN-SUMMARY where provided.
  3. Distinguish actual job rows from diagnostics, partial results, and a valid empty search.
  4. Fetch dataset items through the documented Apify API and validate the exact Actor’s output before mapping fields.

For your agent

Connect only the tools you need.

Use the generic Apify MCP tools to inspect the Actor and run it with explicit input, build, timeout, and cost limits.

https://mcp.apify.com?tools=fetch-actor-details,call-actor,get-actor-run,get-dataset-items,get-key-value-store-record

Ask your agent to inspect the exact Actor schema, use latest, and request a small run only within your authorized budget. Read the run and dataset before claiming results.

Read the public Actor skill ↗