# Changelog of Naukri Scraper — India Jobs, Salary & Company Ratings (`haketa/naukri-scraper`) Actor

- **URL**: https://apify.com/haketa/naukri-scraper/changelog.md
- **Full Actor documentation**: https://apify.com/haketa/naukri-scraper.md

## Changelog

### 0.2.0 — 2026-11-02

- **Verification run (platform, India residential proxy):** broad 5-keyword run (`python developer`, `java developer`, `data analyst`, `sales manager`, `marketing`, `maxTotalItems: 500`) completed **SUCCEEDED but returned 0 items**.
- **Anti-bot status diagnosed:** Naukri's internal `jobapi/v2/search` endpoint now returns an empty payload (`{"totaljobs":0,"list":[]}`, HTTP 200) for all automated clients — tested from datacenter, Apify datacenter proxy, and Apify **residential + country IN**, with Chrome/Firefox fingerprints, full XHR headers, cookie-bootstrapped sticky sessions, and multiple parameter variants. The newer `jobapi/v3/search` returns `406 recaptcha required`. Result listings are gated behind a JavaScript-set anti-bot cookie plus a reCAPTCHA token; the detail endpoints (`simjobs`, `jobapi/v4/job/{id}`) still respond but cannot bootstrap a search.
- **No code change:** the HTTP pipeline is correct for the previous API shape; the block is server-side and was not defeated by any conservative header/param/proxy change, so the code is left unchanged to avoid regression risk. A fix would require a full browser-based rewrite (anti-detect browser, e.g. Camoufox, on India residential IPs) to acquire the JS cookie and pass reCAPTCHA — a large, costly change with uncertain success against a reCAPTCHA-class wall.
- **README:** refreshed to the gold standard — corrected the Output example and field table to match the real dataset schema (18 fields), fixed a stale `searchKeyword`/timestamp in the example, added `customLocations` to the input table, added a dedicated **Pagination & scale** section, and added a prominent, honest note that an **India (`apifyProxyCountry: "IN"`) proxy is required** and that runs may currently return empty result sets due to Naukri's anti-bot changes.
- Metadata (title, description, categories, SEO title/description) reviewed — already complete and accurate; no change.