ats-job-boards-research

ATS job board research

Pull a company's live openings directly from its ATS-hosted career page — Greenhouse, Lever, Workday, Ashby, and 12 more platforms — as normalized JSON from the Crawlora API, no scraping career pages by hand. This is for companies that host their own ATS-backed careers site, not aggregator search (see the job-market-research skill for Indeed/Google Jobs) and not the handful of big-tech-specific skills (Google/Amazon/Apple/Meta/Tesla Jobs).

When to use this skill

• "Pull every open role from ``'s Greenhouse/Lever/Workday board."
• "What is `` currently hiring for on Ashby / SmartRecruiters / Workable?"
• "Get the full description for this one job posting I found on ``'s careers page."
• "Is `` scaling headcount?" — hiring-velocity snapshot from their ATS board.
• "Which ATS does `` use, and what's their board slug?"
• "Diff new postings on ``'s board week over week."

Setup (one-time)

• Get a free Crawlora API key (2,000 credits/mo, no card) at https://crawlora.net.
• export CRAWLORA_API_KEY=sk_your_key_here
• All requests: x-api-key: $CRAWLORA_API_KEY against https://api.crawlora.net/api/v1. Missing/invalid key → 401.

How it works

Every platform follows the same two-call shape: a board/list endpoint (all open postings, often with detail inline) and a job-detail endpoint (one posting's full HTML/text description), both keyed by the company's own slug from its careers URL — not its public brand name. jobs/company-search probes the slug-only platforms for you when you don't already know which ATS a company uses.

Cross-cutting endpoints:

• /jobs/company-search — probe a slug across Greenhouse, Lever, Ashby, SmartRecruiters, Workable, Recruitee, Rippling, Teamtailor, and Pinpoint at once and report where it resolves (Workday excluded — needs tenant+datacenter+site, not a plain slug).
• /jobs/hiring-signals — aggregate any one platform's board into a headcount-growth snapshot (open-role total, department/location/title breakdowns, remote share, roles new in the last 7/30 days) in one call. Pass provider (the ATS name) plus that provider's own slug params.

Full endpoint list, methods, and params: reference/endpoints.md.

Calling the API

# Greenhouse — list a board, then pull one job's full description:
scripts/crawlora.sh /jobs/greenhouse/board token=stripe | jq '.jobs | length'
scripts/crawlora.sh /jobs/greenhouse/job token=stripe id=1234567 | jq '.'

# Lever — list postings filtered by department:
scripts/crawlora.sh /jobs/lever/postings company=netflix department=Engineering | jq '.'

# Workday — needs tenant + datacenter + site from the careers URL:
scripts/crawlora.sh /jobs/workday/board tenant=nike datacenter=wd5 site=nikeinc | jq '.'

# Hiring-velocity snapshot for an Ashby-hosted board:
scripts/crawlora.sh /jobs/hiring-signals provider=ashby org=ramp | jq '.'

Raw curl fallback:

curl -fsS -H "x-api-key: $CRAWLORA_API_KEY" \
  "https://api.crawlora.net/api/v1/jobs/smartrecruiters/postings?company=Visa" | jq '.'

Endpoint reference

See reference/endpoints.md for all 28 endpoints, their HTTP methods, and full param lists.

Examples

• "What is `` hiring for right now?" — if the platform is known, call its board endpoint directly (e.g. /jobs/greenhouse/board token=); otherwise resolve it first with /jobs/company-search slug=.
• "Is `` scaling?" — /jobs/hiring-signals with the right provider + slug params for a velocity summary (new roles in last 7/30 days, department mix), then the raw board endpoint for the role-by-role detail behind it.
• Competitor hiring watch across ATS platforms: pull the ATS boards of 3-4 competitors (mixing Greenhouse, Lever, Workday, whichever each uses) on a schedule and diff new postings between runs.

Notes & limits

• Credits / pay-on-success: billed only on 2xx; free tier 2,000 credits/mo. Key at https://crawlora.net.
• Public data only — public postings/boards; respect each source's terms.
• Security: key lives in CRAWLORA_API_KEY only — never hardcode, query-param, or commit it.
• You need to already know which ATS a company uses and its board slug (not its public brand name) before calling a platform-specific endpoint — use /jobs/company-search to resolve it when you don't. This skill does not do fuzzy company-name lookup or cross-platform discovery beyond that one probe endpoint.
• Results are paginated on most board/list endpoints (limit/offset or page) — walk the full list for a complete pull.