Arbeitsagentur Jobs Scraper: German Job Listings & Salaries
Pricing
from $0.40 / 1,000 jobs
Arbeitsagentur Jobs Scraper: German Job Listings & Salaries
Scrape jobs from the Arbeitsagentur Jobsuche (Bundesagentur für Arbeit): job titles, search links or job links in; title, employer, occupation, place with coordinates, salary, contract, working time, home office and dates out, with the description on request. No login. Misses free.
Jobs from the Bundesagentur für Arbeit's Jobsuche (arbeitsagentur.de), logged out: job titles, keywords, Jobsuche search links or job links in, one row per job out — the reference number and link, title, employer, occupation, the town, postal code, region and coordinates, the working time, home office, contract, the salary the employer states, the start, published and changed dates, and the employer's own link. With includeDetails each row also carries the full description and whether the job is temporary staffing, a private agency's or for applicants with a disability. One free row per search carries the source's count and a verdict: COMPLETE, CAPPED (your limit) or TRUNCATED. No login, no browser.
$0.79 per 1,000 job rows on the Free plan, $0.671 on Starter. A run start is $0.0018. The default input returns about 50 rows for about $0.0413. A value that names nothing is a free row that says why.

Why this Arbeitsagentur jobs scraper
- The Jobsuche's own data, 100 jobs a call. Each search is read from the portal's own job service, the one its app uses, 100 jobs a call until the list ends (the leader's default job, Software Developer, 395 jobs in Germany on 2026-11-12: 4 calls); no page is rendered, so a run of 100 jobs takes a few seconds.
- Searches, search links and job links in one list. A job title or keyword (
Softwareentwickler,Pflegefachkraft), a search link copied from arbeitsagentur.de (its own place, radius and filters are kept), a job link or a bare reference number; the leader'squery(a text, or a JSON list in a text) andstartUrlswork as they are, both at once. - Salary, contract and working time as columns.
salary_min,salary_max,salary_period(hour, month, year) where the employer states pay;contract_type(permanent, fixed term) and its months; full-time, part-time (morning, afternoon, evening, flexible), shift and mini-job flags; home office; the start date. - Every place of a job. The first place in columns (town, postal code, street, region, coordinates, distance from your location) and every place in
locationswhen a job names several (8 of 100 jobs of the default search on 2026-11-12). - Honest runs. A search with no job, an unknown place and a job link the source no longer knows are free rows, each checked on a second connection; every search gets a row with the source's count; a job listed by two searches is charged once a run; the source reads at most 10,000 jobs a search, and a larger search ends TRUNCATED with the count so you can split it.
Who it's for
- Recruiters and staffing agencies — every open job for a role in a region with the employer, the place and the contract, to find hiring companies and leads.
- Labour-market and salary researchers — job counts by occupation, region and offer type over time, with the salaries employers state and the working-time model.
- Job boards and HR tech builders — a daily feed of new German jobs by search, with the reference number to deduplicate and the employer's own link.
Quick start
100 Software Developer jobs in Germany
{"queries": ["Software Developer"],"maxItems": 100}
Care jobs within 25 km of Berlin, published this week, with the description
{"queries": ["Pflegefachkraft"],"location": "Berlin","radius": 25,"publishedSince": 7,"includeDetails": true,"maxItems": 200}
Apprenticeships in Hamburg from a Jobsuche search link
{"queries": ["https://www.arbeitsagentur.de/jobsuche/suche?angebotsart=4&was=Koch&wo=Hamburg&umkreis=25"],"maxItems": 100}
Switching from blackfalcondata/arbeitsagentur-jobs-feed?
Your input works as it is. Each name in the left column is read as this Actor's own field and the run log says so (read <theirs> as <ours>); a name marked not read is ignored with a note. A value in the other Actor's shape, an object list or its own date form, is read too.
| blackfalcondata/arbeitsagentur-jobs-feed | this Actor | what to know |
|---|---|---|
query | queries | A job title or keyword; a JSON list in the text, or a list, gives one search each. |
location | location | A town, a postal code or a federal state. |
maxResults | maxItems | The cap of the whole run; 0 = no cap. |
startUrls | queries | Job links and search links are read in the same list (joined to queries when both are given). |
mode | not read | Incremental mode is not read: every run returns the jobs the source lists now; compare id and modified_at between runs. |
includeDetails | includeDetails | The same switch: the description and the staffing, agency and disability flags, one more call a job. |
radius | radius | Kilometres around the location; the default here is the portal's 25. |
contractType | not read | Not an input: contract_type (permanent, fixed_term) is on every row; a search link with the portal's contract filter keeps it. |
jobType | jobType | ARBEIT, AUSBILDUNG, SELBSTAENDIGKEIT and PRAKTIKUM are read as job, apprenticeship, self_employment and internship. |
workType | not read | Not an input: the full-time, part-time, shift, mini-job and home-office flags are on every row; a search link keeps the portal's filter. |
bundesland | not read | Give the state as location (Bayern, Nordrhein-Westfalen). |
remoteOnly | not read | Not an input: filter the rows on remote_possible. |
disabilityOnly | not read | Not an input: with includeDetails, filter the rows on for_disabled_applicants. |
careerChange | not read | Not an input: filter the rows on career_change. |
sortBy | not read | Not read: the source's own order (relevance) is kept, with rank on each row. |
publishedSince | publishedSince | Only jobs published in the last N days; 0 = any date. |
includeTemporaryWork | not read | Not an input: with includeDetails, filter the rows on temporary_staffing. |
employer | not read | Not an input: put the employer's name in the search words, or filter on employer. |
skipReposts | not read | Not read: a job is charged once a run by its reference number. |
emitUnchanged | not read | Incremental mode is not read. |
emitExpired | not read | Incremental mode is not read. |
compact | not read | Not read: every row has every column, null where the source has no value. |
descriptionMaxLength | not read | Not read: the description comes whole (with includeDetails). |
descriptionFormat | not read | Not read: the source has one plain-text description. |
excludeEmptyFields | not read | Not read: every row has every column, so the dataset's schema holds. |
telegramToken | not read | Notifications are not sent: use an Apify integration or webhook on the run. |
telegramChatId | not read | Notifications are not sent. |
discordWebhookUrl | not read | Notifications are not sent: use an Apify integration or webhook on the run. |
slackWebhookUrl | not read | Notifications are not sent: use the Apify Slack integration. |
whatsappPhoneNumberId | not read | Notifications are not sent. |
whatsappAccessToken | not read | Notifications are not sent. |
whatsappTo | not read | Notifications are not sent. |
webhookUrl | not read | Not read: set a webhook on the run in Apify Console. |
webhookHeaders | not read | Not read: set a webhook on the run in Apify Console. |
notificationLimit | not read | Notifications are not sent. |
notifyOnlyChanges | not read | Notifications are not sent. |
appConnector | not read | Not read: export the dataset with an Apify integration. |
mcpIssueTeam | not read | Not read. |
What differs:
queryandstartUrlsare one list here (queries); the leader's two are joined into it, both at once. A JSON list inqueryis read as several searches.maxResultsismaxItems, the cap of the whole run, with the same default (100); 0 = no cap beyond your charge limit.radiusdefaults to 25 km (the portal's own default) where the leader's is 0; give 0 for the place itself.- Incremental mode, the notifications and the connectors are not read: schedule the Actor and compare
idandmodified_atbetween runs. - The filters for contract, working time, state, home office, disability, career change, employer and temporary work are not inputs: every row carries each of them as a column, so filter the dataset; a search link copied from the portal keeps its own filters.
- Text-mined e-mails, phones and social links are not read: they are a contact person's data. The
description(withincludeDetails) carries what the employer wrote.
Output
One row per job (and one free summary row per search or job link):
| field | meaning |
|---|---|
record_type | job on a job row (charged); target on a search's or a job link's free summary row. |
status | ok, or why an input gave no job: no_results, unknown_location, not_found, invalid_input, failed, not_read. |
input | The value you gave that this row answers (for a job, the search that listed it). |
id | The job's reference number at the Bundesagentur für Arbeit (Referenznummer). |
url | The job's page on arbeitsagentur.de. |
title | The job's title as the employer wrote it. |
employer | The employer's name (Firma). |
occupation | The job's main occupation in the source's occupation list (Hauptberuf). |
all_occupations | Every occupation the source files the job under. |
job_type | The offer type: job, apprenticeship, self_employment or internship. |
city | The town of the job's first place. |
postal_code | The postal code of the job's first place. |
street | The street of the job's first place, where the employer gives it. |
region | The federal state of the job's first place (BAYERN, BERLIN, …). |
country | The country of the job's first place (DEUTSCHLAND, or another where the job is abroad). |
latitude | The latitude of the job's first place. |
longitude | The longitude of the job's first place. |
locations | Every place of the job (town, postal code, street, region, country, coordinates) when it names more than one; null when it names one. |
distance_km | Kilometres from your location to the job, when a location was searched. |
full_time | True for a full-time job. |
part_time | True when any part-time model is offered. |
part_time_morning | Part time in the morning. |
part_time_afternoon | Part time in the afternoon. |
part_time_evening | Part time in the evening. |
part_time_flexible | Part time with flexible hours. |
night_weekend_shift | Shift, night or weekend work. |
mini_job | A mini job (geringfügige Beschäftigung). |
remote_possible | True when the employer offers home office; null on a job read from its link (the job's own record does not say). |
remote_type | How the home office is given: by_agreement or share_given. |
contract_type | permanent or fixed_term; null when the employer does not say. |
contract_months | The months of a fixed-term contract, where given. |
salary_min | The lowest pay the employer states, in salary_currency per salary_period; null when no pay is stated. |
salary_max | The highest pay the employer states (equal to salary_min for a fixed salary). |
salary_currency | EUR when a salary is stated. |
salary_period | hour, month or year: what the salary figures are paid for. |
salary_kind | range or fixed. |
start_date | The first day of the job (the earliest start). |
published_date | When the job was published in its current form. |
first_published_date | When the job was first published. |
modified_at | When the job was last changed at the source. |
career_change | True when the job is open to career changers (Quereinsteiger). |
external_url | The employer's or the original board's own page of the job, where it came from there. |
employer_reference | The employer's own reference of the job (Chiffrenummer), where given. |
employer_hash | The source's stable key of the employer: equal on every job of one employer account. |
details_status | not_asked (no includeDetails), served, gone (the job left the source between the list and its record) or failed. |
description | The job's full description text (with includeDetails, or for a job link). |
temporary_staffing | True for temporary staffing (Arbeitnehmerüberlassung, Zeitarbeit); with includeDetails. |
private_agency | True when a private placement agency posted the job; with includeDetails. |
for_disabled_applicants | True when the job is offered to applicants with a severe disability; with includeDetails. |
partner_name | The source's alliance partner that supplied the job, where one did; with includeDetails. |
partner_url | The alliance partner's site; with includeDetails. |
rank | On a job row: its place in the search's order (1 = first). |
total_count | On a search's row: the jobs the source counts for the search. |
pages_read | On a search's row: the source's pages of 100 jobs read. |
verdict | On a target row: COMPLETE (read to the end), CAPPED (stopped at maxItems), TRUNCATED (the source's 10,000-job depth, or a list that ended under its count) or PARTIAL (stopped: note says why). |
rows_delivered | On a target row: the job rows it delivered. |
note | Why a target row has the status or the verdict it has. |
scraped_at | When this row was read (UTC). |
Example record:
{"record_type": "job","status": "ok","input": "software developer","id": "16752-r2580-S","url": "https://www.arbeitsagentur.de/jobsuche/jobdetail/16752-r2580-S","title": "Software Developer* Cloud-Technologien","employer": "inovex GmbH","occupation": "Softwareentwickler/in","all_occupations": ["Softwareentwickler/in","Cloud-Architect","DevOps Engineer"],"job_type": "job","city": "Berlin","postal_code": null,"street": null,"region": "BERLIN","country": "DEUTSCHLAND","latitude": 52.4926833,"longitude": 13.4025753,"locations": null,"distance_km": 3,"full_time": true,"part_time": false,"part_time_morning": false,"part_time_afternoon": false,"part_time_evening": false,"part_time_flexible": false,"night_weekend_shift": false,"mini_job": false,"remote_possible": false,"remote_type": null,"contract_type": null,"contract_months": null,"salary_min": 60000.0,"salary_max": 75000.0,"salary_currency": "EUR","salary_period": "year","salary_kind": "range","start_date": "2026-11-02","published_date": "2026-03-23","first_published_date": "2025-05-22","modified_at": "2026-11-02T07:31:25.468","career_change": false,"external_url": null,"employer_reference": "r2580","employer_hash": "9xPxVzI0AAN8g51FCj6infUxfay0yhdWR1aBH35sNw0=","details_status": "not_asked","description": null,"temporary_staffing": null,"private_agency": null,"for_disabled_applicants": null,"partner_name": null,"partner_url": null,"rank": 3,"total_count": null,"pages_read": null,"verdict": null,"rows_delivered": null,"note": null,"scraped_at": "2026-11-12T19:52:34Z"}
Pricing
Pay for the jobs a run delivers, plus one run start; no platform usage is billed to you. The job price falls with your Apify plan.
| event | unit | Free | Starter | Scale | Business |
|---|---|---|---|---|---|
Job (job) | per 1,000 | $0.79 | $0.671 | $0.553 | $0.395 |
Run start (start) | per run | $0.0018 | $0.0018 | $0.0018 | $0.0018 |
A job is $0.79 per 1,000 on the Free plan ($0.00079 a job), with the description and the staffing, agency and disability flags of includeDetails included at the same price. Each run that delivers a charged job also pays the run start, $0.0018 once; a run that delivers no charged job pays nothing at all. So 100 jobs of one search cost $0.0808 on the Free plan, and several searches in one run share one start. The free target row of every search, a value that names nothing, a search with no job and a job the source no longer knows are never charged, and a job listed by two searches is charged once a run. Set a charge limit on the run and it stops before it would pass it.
Usage patterns
- Searches, search links and job links —
queriestakes one value a line: job titles or keywords (Softwareentwickler,Software Developer,Pflegefachkraft), a search link copied from arbeitsagentur.de/jobsuche (itswas,wo,umkreisand the portal's filters are kept), a job link (/jobsuche/jobdetail/<number>) or a bare reference number (12704-54RY6AW2I2B1TUGU-S).location,radius,jobTypeandpublishedSinceapply to every search that does not name its own. - What a row is —
record_type=job: one job, withrankits place in the search's order andinputthe search that listed it.record_type=target: the free row of one search or job link —total_count(the jobs the source counts for the search),pages_read,rows_delivered, theverdictand thestatuswith itsnote. - How far a run goes —
maxItems(default 100; 0 = no cap) caps the jobs of the whole run. The source serves 100 jobs a call in about 0.1 to 0.6 s and reads at most 10,000 jobs a search: a search the source counts larger ends TRUNCATED with its count, so split it bylocation,radiusorpublishedSince. Every job in Germany: 790,838 on 2026-11-12.
Input configuration
| field | type | default | what it does |
|---|---|---|---|
queries | array | [] | Each value is read on its own: a job title or keyword (Softwareentwickler, Pflegefachkraft, Software Developer), a Jobsuche search link copied from arbeitsagentur.de (its own place, radius and filters are kept), a job link (/jobsuche/jobdetail/ |
location | string | "" | A town, a postal code or a federal state (Berlin, 80331, Bayern), for every search that does not name its own. Empty: all of Germany. |
radius | integer | 25 | Kilometres around the location (the portal's default is 25). 0 = the place itself. Used only with a location. |
jobType | all / job / apprenticeship / self_employment / internship | "all" | Which offers to read: every type, jobs, apprenticeships and dual studies, self-employment, or internships and trainee posts. |
publishedSince | integer | 0 | Only jobs published in the last N days (1 = the last day, 7 = the last week). 0 = any date. |
includeDetails | boolean | false | One more call a job: the full description text, whether the job is temporary staffing (Zeitarbeit), from a private placement agency or for applicants with a disability, and the alliance partner. The price per job is the same. |
maxItems | integer | 100 | The most jobs of the whole run, each charged once; 0 = no cap beyond your charge limit. |
feedback | string | A note to the people who maintain this Actor: what you needed and did not get. For example a missing field, a wrong or empty result, or the Actor you switch from or to, and why. The run does not read it: the rows and the charges stay the same. A person reads every note. |
Errors
A run does not fail because of your input. When an input cannot be used — a date that is not a date, an id in the wrong form, two settings that contradict each other — the run ends Succeeded, charges nothing, and says what to change:
- the status message starts with
INPUT REJECTED; - the dataset holds one row,
{"error": true, "code": "…", "message": "…"}, and no results; - the key-value store record
ERRORholds the same object.
From code, check error on the first row before reading results.
A value that names no search or job (a link to another site, an arbeitsagentur.de page that is not the Jobsuche) is not an error: it gets one free target row with status invalid_input that says why, and the run goes on with the others. A run that ends Failed is a fault on our side, never your input; it charges nothing, and we are alerted.
FAQ
Which status is which?
A target row's status is ok (jobs served), no_results (the source lists no job for the search; checked on two connections), unknown_location (the source does not know the place), not_found (the job link's job is gone; checked on two connections), invalid_input (not a search or a job, or a search the source refused), failed (the source refused every try; free) or not_read (left at maxItems, the charge limit or the run's time limit). A job row's status is ok.
Does it need an account or an API key?
No. The jobs are read from the Jobsuche's public job service as a visitor who is not logged in. Nothing that needs a login is read.
Why is the salary empty on most jobs?
The source shows a salary only where the employer states one (24 of the 100 jobs of the default search on 2026-11-12). salary_kind says whether it is a range or a fixed salary and salary_period whether it is an hourly, monthly or yearly figure; an empty salary is null, never 0.
Are there e-mails and phone numbers?
No contact columns: an application contact is usually a person, so this Actor does not mine them. With includeDetails the description carries the job text as the employer wrote it, and external_url the employer's own job page where the job was posted from another board.
Do I need Apify Proxy or my own proxy?
No. The Actor uses its own connections, at no extra cost to you.
Integration
JavaScript
import { ApifyClient } from 'apify-client';const client = new ApifyClient({ token: 'YOUR_TOKEN' });const run = await client.actor('hyperbach/arbeitsagentur-jobs-scraper').call({"queries": ["Software Developer"], "maxItems": 100});const { items } = await client.dataset(run.defaultDatasetId).listItems();
Python
from apify_client import ApifyClientclient = ApifyClient('YOUR_TOKEN')run = client.actor('hyperbach/arbeitsagentur-jobs-scraper').call(run_input={'queries': ['Software Developer'], 'maxItems': 100})items = client.dataset(run['defaultDatasetId']).list_items().items
CLI
$apify call hyperbach/arbeitsagentur-jobs-scraper --input '{"queries": ["Software Developer"], "maxItems": 100}'
REST
curl -X POST "https://api.apify.com/v2/acts/hyperbach~arbeitsagentur-jobs-scraper/run-sync-get-dataset-items?token=YOUR_TOKEN" \-H 'Content-Type: application/json' -d '{"queries": ["Software Developer"], "maxItems": 100}'
AI agents and MCP
Apify's MCP server offers this Actor as a tool to Claude, Cursor, VS Code and any other MCP client. Connect https://mcp.apify.com?tools=hyperbach/arbeitsagentur-jobs-scraper and ask, for example: "Get 100 Software Developer jobs in Berlin from the Arbeitsagentur Jobsuche with employer, location, salary, contract type and publication date". The agent fills the input from the table above and reads the rows from the run's dataset. A field under the name another Actor gives it is read as this Actor's own: query, startUrls, keywords, keyword as queries; maxResults, maxJobs, limit, maxRecords as maxItems; wo, where, city, place as location; umkreis, distance, radiusKm, distanceKm as radius (and more). An agent or a person can add feedback to any run and say what was needed and not delivered: a missing field, a wrong or empty result, the Actor switched from or to, and why. The run returns the same rows at the same price, and a person reads every note.
n8n, Make and Zapier
The Apify node in n8n, the Apify app in Make and the Apify integration in Zapier run this Actor on a schedule or from a trigger and read its dataset. A webhook on the run's end (Integrations on the Actor page) sends the rows on to your own endpoint. Nothing to install on our side.
Support
Disagree with a charge? Write to apify@hyperbach.com with the run ID. We will ask Apify to compensate you, and the charge comes back to your Apify account as platform credit.
Did it do the job? A short review on the Store, at any rating, helps the next buyer choose.
See also
- Upwork Jobs History — every Upwork job posting since November 2024 by period, category and skill, with the client's record as it stood on the day of posting.
- LinkedIn Jobs Scraper — LinkedIn job listings for keywords and a location, with the applicants count, the salary block and the company's facts.
- Greenhouse Jobs Scraper — the hiring of major tech companies from their own Greenhouse boards, kept as roles open and close.
- Remote Jobs Scraper — every remote job from seven boards and ATSs in one deduplicated feed, kept after the postings close.
- Y Combinator Jobs Scraper — every open job of every YC company on Work at a Startup, newest first, by keyword, company or link: the description, salary and equity, the role, posting age, remote, visa, the batch and the founders.
- Internshala Scraper — Internshala internships and fresher jobs in India by keyword, category, city or link, newest first: the stipend or salary as numbers, applicants, openings, the apply-by date, the exact post time, skills and perks.
This page is generated from the Actor's schemas and a live sample — it cannot describe a field the Actor does not have.