# Changelog of Multi-ATS Jobs Scraper (Greenhouse, Lever, Ashby) (`devilscrapes/multi-ats-jobs-scraper`) Actor

- **URL**: https://apify.com/devilscrapes/multi-ats-jobs-scraper/changelog.md
- **Full Actor documentation**: https://apify.com/devilscrapes/multi-ats-jobs-scraper.md

## Changelog

### 0.2 — 2026-07-20

Auto-detect ATS from a bare company slug or careers URL — `companies` now
accepts `str | CompanyEntry`, so a legacy `{atsType, companySlug}` object,
a bare slug (concurrent 3-way probe), or a full careers/ATS URL (regex
extraction, zero probes) can all be mixed in one run.

`ResultRow` schema changes: `source_ats` → `ats`, `description` →
`description_html`, dropped `employment_type`, added `team`, `salary_min`,
`salary_max`, `salary_currency` (Ashby only — Greenhouse and Lever don't
expose structured salary). New input filters: `departmentFilter`,
`locationFilter`, `remoteOnly`, applied before `maxItemsPerCompany`.

Migrated the HTTP layer from `httpx` to `curl-cffi` with browser
impersonation and rotation on `403`/`429` (ADR-0002 compliance fix — v0.1's
`httpx` client was an undocumented violation; none of the three ATS hosts
are on the `httpx` allowlist). Retry status set widened to
`{408, 429, 500, 502, 503, 504}`.

Fixed a stale README/promo.json claim: Ashby's job-board API is REST
(`api.ashbyhq.com/posting-api/job-board`), not GraphQL. Removed a
brand-voice violation from `promo.json` ("no anti-bot, fast and reliable").

Cloud QA: pending (T19, not yet run as of this entry).

### 0.1 — 2026-06-07

Initial release. Greenhouse, Lever, and Ashby scrapers via public JSON APIs. Normalised `ResultRow` schema with title, location, department, employment type, remote flag, and description. Pay-Per-Event pricing at $0.0015/result.

Cloud QA: PASS — runId `iedkkN08dIydfE7BJ`, 9 records, PPE charges `actor-start=1, result=9`.