ATS Job Scraper — Greenhouse, Lever, Workday & more

Pull open job listings straight from company career sites — no LinkedIn, no Indeed, no proxies, no risk. This actor reads the same public, unauthenticated APIs that power the career pages companies embed on their own websites, across six applicant-tracking systems:

ATS	Token format	Example
Greenhouse	greenhouse:<board>	greenhouse:stripe
Lever	lever:<site>	lever:palantir
Workday	workday:<host>/<site>	workday:nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite
SmartRecruiters	smartrecruiters:<company>	smartrecruiters:visa
Recruitee	recruitee:<subdomain>	recruitee:gorgias
Workable	workable:<account>	workable:turnitin

What you get

One dataset item per job: company, ats, title, location, remote, employment_type, department, salary, posted_at, apply_url, description_text. Salary appears where the company publishes it (Lever ranges, and increasingly others under pay-transparency laws).

No personal data, by design. The schema has no recruiter names, no emails, no phone numbers — and description text is scrubbed of both before it reaches the dataset. Use it for job-market analytics, hiring-signal tracking, competitor watching, or feeding an agent, without inheriting a compliance problem.

Where do I find a company's token?

Open the company's careers page and look at the URL of a job listing:

• boards.greenhouse.io/stripe/... → greenhouse:stripe
• jobs.lever.co/palantir/... → lever:palantir
• nvidia.wd5.myworkdayjobs.com/en-US/NVIDIAExternalCareerSite/... → workday:nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite
• jobs.smartrecruiters.com/Visa/... → smartrecruiters:visa
• <name>.recruitee.com/o/... → recruitee:<name>
• apply.workable.com/<name>/... → workable:<name>

Input example

{
    "companies": ["greenhouse:stripe", "lever:palantir"],
    "max_jobs_per_company": 500,
    "include_description": true
}

Up to 100 companies per run. Failed companies (moved ATS, typo, empty board) come back as type: "company-error" items so your pipeline can branch — the run never dies because one token is stale.

Pricing

Pay-per-event: a small start fee, a micro-fee per company processed, and a per-job fee. No subscription, no minimum. A 500-job run costs well under a dollar.

Notes & limits

• Tokens go stale. Companies switch ATS vendors; a token that worked last month can 404. That is what the company-error items are for — prune and replace.
• Workday listing data is thinner than the others (no posted date in the listing API, location as display text). Detail-page fetching is not included in v0.1.
• Only these six ATSes. No LinkedIn, no Indeed, no Glassdoor — deliberately: those require scraping platforms that prohibit it. This actor only touches endpoints companies expose publicly on purpose. Ashby is excluded because its robots.txt disallows its API.
• Descriptions are plain text (HTML stripped). If you need the original HTML formatting, tell us and it becomes v0.2.

Other actors by amanatools

• PDF Text Extractor — tables and text from PDF files into clean JSON rows
• PDF OCR Extractor — scanned and image-only PDFs into searchable text
• Data Cleaner — messy CSV, Excel and JSON into clean, typed data
• Doc to Markdown — DOCX, PDF and web pages into clean markdown