Use case · JobsRadar

One job postings API for Greenhouse, Lever and the rest

Search open roles across company boards on Greenhouse, Lever, Ashby, SmartRecruiters, Personio and Pinpoint in one call, deduplicated and in one normalized shape.

The job

Most companies publish their open roles through an applicant tracking system, and the big ones each expose a public board feed: Greenhouse, Lever, Ashby, SmartRecruiters, Personio, Pinpoint. Each has its own URL scheme, its own field names and its own idea of what a location or a department is.

Building a job search across companies means calling each board, mapping each schema to one, and removing the postings that appear twice. That is the part JobsRadar does.

Why an agent needs a tool for it

An agent helping someone find a role, or tracking a market, should not have to know which tracking system each company uses. search_jobs takes title keywords, company names and filters, and returns one list where every posting has the same fields and an apply URL.

The answer also says how fresh it is (data_as_of), how many boards were searched and which ones failed or were too large to read in full, so the agent can say what it did not see.

The calls

One tool call.

Live production call2026-10-01

Agent calls search_jobs

{
  "name": "search_jobs",
  "arguments": {
    "title_keywords": [
      "data engineer",
      "backend"
    ],
    "companies": [
      "Spotify",
      "Datadog",
      "Figma"
    ],
    "limit": 3
  }
}

JobsRadar answers

{
  "count": 3,
  "data_as_of": "2026-10-01T17:45:15.000Z",
  "boards_searched": 3,
  "boards_live": 0,
  "boards_failed": [],
  "boards_truncated": [],
  "postings_seen": 684,
  "duplicates_removed": 36,
  "jobs": [
    {
      "id": "lever:spotify:65d6caca-6d7c-4049-8256-a128e0e7249e",
      "title": "Backend Engineer - Music",
      "company": "Spotify",
      "location": "New York, NY",
      "remote": true,
      "department": "Engineering",
      "postedAt": "2026-09-29T13:23:55.240Z",
      "applyUrl": "https://jobs.lever.co/spotify/65d6caca-6d7c-4049-8256-a128e0e7249e",
      "provider": "lever"
    },
    {
      "id": "greenhouse:figma:6178851004",
      "title": "Data Engineer Intern (2027)",
      "company": "Figma",
      "location": "San Francisco, CA • New York, NY",
      "remote": false,
      "postedAt": "2026-09-23T19:30:27.000Z",
      "applyUrl": "https://boards.greenhouse.io/figma/jobs/6178851004?gh_jid=6178851004",
      "provider": "greenhouse"
    }
  ]
}

A Lever posting and a Greenhouse posting in the same shape. provider says where each came from; the id is stable across calls. Shortened for this page: jobs shortened from 3 to 2.

Limits worth knowing

Search covers the tracked directory, read with list_supported_companies. For a company that is not tracked, use get_company_jobs with its board slug.
Workable boards refuse this hosted service and work through the Apify Actor instead.
Locations are the text each company published, not geocoded places. The location filter is a substring match on that text.

Pricing

1 credit for the calls on this page.

Tool	Credits per call
search_jobs	1

Credits are prepaid and shared across every server. Starter is $10 for 1,000 credits, packs never expire, and 100 credits are free without a card. Full pricing.

Prefer to run it without an MCP client? JobsRadar is also an Apify Actor, billed by Apify at $0.90 per 1,000 postings.

Install

Connect JobsRadar to your agent.

A remote MCP server over Streamable HTTP. Get a free key, swap it in, and the calls above work as written, with answers reflecting the data on the day you call.

EndpointStreamable HTTP

https://jobsradar.agenttoolworks.com/mcp

Connect it, Claude Code

claude mcp add --transport http jobsradar \
  https://jobsradar.agenttoolworks.com/mcp \
  --header "Authorization: Bearer atw_live_your_key"

Setup for Claude Desktop, Cursor and Windsurf is in the docs.