GovernmentJobs Jobs Search Scraper

Extract rich, source-grounded public job records from GovernmentJobs.com. The Actor discovers official listing results, fetches each numeric GovernmentJobs detail page, verifies the canonical URL and posting identity, and writes only complete records to the default dataset.

What it captures

Each completed record can include more than 20 populated values, including the job ID and number, title, canonical and application URLs, search position and result count, employer name/profile/address/phone/website, location hierarchy, coordinates when published, employment and remote labels, salary text and normalized amount/currency/unit, opening and closing dates, department, area, bargaining unit, licensing, safety-sensitive and on-call flags, responsibilities, qualifications, requirements, skills, benefits, screening questions, job conditions, disclaimers, sectioned HTML/text descriptions, structured-data availability, and hashed HTTP response receipts.

The output is deliberately normalized rather than a raw page dump. Published fields that are absent for a particular posting are omitted, and no fixture, debug, proxy, retry, timeout, concurrency, or raw-payload switch is exposed in the public input schema.

Input

The supported public inputs are:

• mode: search, searchMultiple, single, multiple, or startUrls.
• query, location, and category for the public listing search.
• searchQueries for bounded multi-query searches.
• startUrls, jobUrl, or jobUrls for official listing/detail URLs.
• maxItems, maxCandidates, and maxPages for bounded collection.

Example:

{
  "mode": "search",
  "query": "software engineer",
  "location": "",
  "maxItems": 3,
  "maxCandidates": 10,
  "maxPages": 1
}

Output

Records use the strict schema in .actor/dataset_schema.json. responseReceipts.detail contains the requested URL, final URL, HTTP status, content type, byte count, SHA-256 body hash, attempt count, and duration. This makes the source and completeness of each record auditable without publishing the original structured payload.

You can download the dataset in JSON, CSV, Excel, HTML, or other Apify-supported formats.

Local validation

RUN_SUMMARY.status and RUN_HEALTH.status report the Apify platform outcome (SUCCEEDED or FAILED); resultStatus reports dataset completeness (COMPLETE, LIMITED, SKIPPED, DEFERRED, or FAILED). Only awaited successful Dataset writes are counted. A stored verified prefix remains successful/limited after a later write or runtime error. Diagnostics and placeholder records are never Dataset rows. Ambiguous zero-result parses are DEFERRED; an explicit official access barrier stops further requests and reports SKIPPED.

Run npm test, npm run check, npx apify-cli validate-schema, and npm run validate after an ordinary local run. The validator checks unique IDs and URLs, complete detail verification, non-empty nested values, the more-than-20 meaningful-field threshold, HTTP-200 response receipts, and the run summary. The Actor uses ordinary direct HTTPS only; it does not attempt to bypass an access barrier.

Responsible use

Collect only public job information and follow GovernmentJobs terms, robots directives, and applicable law. Do not use the Actor for authenticated pages, candidate personal data, paywall circumvention, or CAPTCHA/security-bypass activity. If the site explicitly blocks ordinary access, stop and record the actor as skipped with the observable evidence.