HireHive Jobs Monitor & Scraper

Read jobs from known public HireHive company boards. Get role text, dates and links, then compare full snapshots to see new, updated and closed jobs. Add a board such as ksn and press Start.

Each run costs $0.027 to start + $0.0036 per saved job or change. Empty scans and checks with no changes still have the start fee. A failed or capped scan is not a safe monitoring baseline.

Contents

• What it does
• Data you get
• Quick start
• Input fields
• Input examples
• Output example
• Output fields
• Pricing and limits
• Use cases
• Monitoring
• API and integrations
• Limits and partial scans
• Troubleshooting
• FAQ
• Related Actors
• Support

What it does

The Actor reads the public Jobs API on a HireHive board. It checks every page and the total job count. It keeps each job once by its HireHive ID. It reads up to four known company boards in one run.

It does not search for companies. Use a company slug from a real HireHive careers link. It does not open candidate accounts, collect resumes or send applications. No HireHive login, API key or proxy setup is needed.

Data you get

You get the title, company, location, job type, category, language, published date and job link. Salary is returned as the source text when the company gives it. Role text is plain text. Missing source fields are null.

In change mode, changeType is new, updated or closed. Updates include changes to role text. A closed job means that its ID left a complete board snapshot. It does not prove why the employer removed it.

Dataset has saved job rows. Key-value store has OUTPUT, RUN_SUMMARY, SNAPSHOT and CHANGES. SNAPSHOT keeps the full scanned list before title filters, including role text even when you hide it in Dataset.

Quick start

1. Open Input.
2. Add ksn to HireHive boards. Leave the other values at their defaults.
3. Press Start.
4. Open Dataset. Look for Contact Centre Agent, a Dublin location and its HireHive link. On the check date, KSN had six jobs. The live count can change.
5. Open RUN_SUMMARY in Key-value store. Check that status is complete and counts.delivered equals the Dataset count.

Input fields

Key	Type	Required / default	Allowed values	What it changes	Recommendation
boards	string array	No; ["ksn"]	1–4 entries; lowercase tenant slug or HTTPS tenant root URL	Reads known company boards. Duplicate entries are read once.	Copy the root from the company careers page. UI prefill is also ksn.
keywords	string array	No; []	Up to 20 non-empty phrases, 1–100 characters each	Keeps titles that contain any phrase, ignoring case.	Start empty; filtering happens after the full scan.
maxResults	integer	No; 53	1–53	Caps saved rows, including closed jobs.	Keep 53 for monitoring. A cap that drops rows makes the snapshot unsafe.
includeDescription	boolean	No; true	true, false	Shows role text in Dataset, or returns descriptionText: null.	Leave true for a job feed. It does not change source requests or change detection.
incremental	boolean	No; false	true, false	Returns changes only. The first check returns all jobs as new.	Set true when comparing snapshots.
previousRunId	string	No; ""	Empty or a 17-character Apify run ID	Uses a previous complete, successful run from this Actor, with exactly the same boards. Requires incremental=true.	Save the last complete run ID. Do not use a partial, missing or failed run.

Do not pass null or unknown keys. An empty board list is invalid. A full job link, custom domain, HTTP URL, credentials, URL query or fragment is not a supported board input. Examples of valid values are addo and https://ksn.hirehive.com/.

Filters affect saved rows. They do not remove jobs from the full snapshot. You may change a title filter between checks, but change mode still compares the full board: it does not mark unchanged jobs as new just because a new filter matches them.

Input examples

Read two boards

{
  "boards": ["ksn", "addo"],
  "includeDescription": true,
  "maxResults": 53
}

These boards had seven jobs in the check. Look for KSN jobs and ADDO's Business Analyst. Counts may change.

Check only changed jobs

{
  "boards": ["ksn", "addo"],
  "incremental": true,
  "previousRunId": "tio7rFP46uDwcraEo",
  "maxResults": 53
}

Replace the sample ID with your own successful complete run ID from example 1. The displayed ID is a format example; it is not a shared baseline. If nothing changed, Dataset is empty and SNAPSHOT.complete is true.

Keep analyst titles

{
  "boards": ["addo"],
  "keywords": ["analyst"],
  "includeDescription": false
}

The saved role text is null. The full role text remains in SNAPSHOT for comparison. The price includes the start fee even if no title matches.

Output example

The example below comes from the ADDO board in example 1. Role text is shortened here for reading; the Dataset row has the complete supported text.

{
  "schemaVersion": 1,
  "ats": "hirehive",
  "org": "addo",
  "company": "ADDO",
  "companyTotalJobs": 1,
  "jobId": "79402",
  "globalId": "hirehive:addo:79402",
  "title": "Business Analyst",
  "department": null,
  "employmentType": "FullTime",
  "language": "en",
  "languageLabel": "English",
  "location": "12/2 A, Sector XX, Phase 3, DHA, Lahore, Pakistan",
  "locations": ["12/2 A, Sector XX, Phase 3, DHA, Lahore, Pakistan"],
  "address": {
    "city": "12/2 A, Sector XX, Phase 3, DHA, Lahore",
    "country": "Pakistan",
    "countryCode": "PK"
  },
  "salaryText": null,
  "descriptionText": "Responsibilities We are looking for a business analyst (BA) who will primarily serve as a liaison between our clients and project resources.",
  "publishedAt": "2021-05-19T10:09:46.000Z",
  "jobUrl": "https://addo.hirehive.com/business-analyst-122-a-sector-xx-phase-3-dha-lahore-L0BpOZ",
  "status": "complete",
  "changeType": null,
  "scrapedAt": "2026-10-04T09:01:55.579Z",
  "contentHash": "1ad053a5ca09de25dafa201a36482edcb382bd814de520dea590fff94a4324b4"
}

contentHash is a 64-character SHA-256 value in real output. Read it from the actual row when comparing data.

Output fields

All fields below are present in every Dataset row. A null value means that the source did not give the field, or that a setting hid it.

Field	Type / null	Example	Meaning
schemaVersion	integer; not null	1	Output contract version.
ats	string; not null	hirehive	Source platform.
org	string; not null	addo	Board slug.
company	string; not null	ADDO	Name from the board page.
companyTotalJobs	integer; not null	1	Total published jobs reported by the source at scan time. Closed rows keep their previous value.
jobId	string; not null	79402	Numeric HireHive ID as a string.
globalId	string; not null	hirehive:addo:79402	Stable platform + board + ID key. Use this as your database key.
title	string; not null	Business Analyst	Public job title.
department	string or null	Finance	Source category.
employmentType	string or null	FullTime	Source job type code; no guess from title.
language	string or null	en	Base language code from the source.
languageLabel	string or null	English	Source language name.
location	string or null	Dublin, Ireland	Source city/location joined with country.
locations	string array; can be empty	["Dublin, Ireland"]	This API gives one place; empty when absent.
address	object; not null	{"city":"Dublin","country":"Ireland","countryCode":"IE"}	Source place fields.
address.city	string or null	Dublin	Source location text. It may be a region or a full address.
address.country	string or null	Ireland	Source country name.
address.countryCode	string or null	IE	Source country code.
salaryText	string or null	Competitive	Raw salary text, when published. No currency conversion.
descriptionText	string or null	Responsibilities ...	Plain role text; null when includeDescription=false. Email and phone patterns and personal LinkedIn links are removed. No candidate data is fetched.
publishedAt	ISO date string or null	2021-05-19T10:09:46.000Z	API publishedDate, converted from Unix seconds. Not the scan date.
jobUrl	string; not null	https://addo.hirehive.com/...	Public job page. Tracking query is removed.
status	string; not null	complete	The row was parsed. Read the run summary for full scan status.
changeType	string or null	updated	null in normal mode; new/updated/closed in change mode.
scrapedAt	ISO date string; not null	2026-10-04T08:48:34.594Z	Time this run started reading data.
contentHash	string; not null	64 hex characters	Stable hash of public content. It excludes scan time, change type and board job count.

Key-value store records

Record	Fields and purpose
OUTPUT	Same run summary as RUN_SUMMARY; includes status, counts, boards, errors and storage IDs.
RUN_SUMMARY	schemaVersion integer; status complete/partial/not_found/error; complete boolean; counts object; boards array; errors array; datasetId string or local null; snapshotKey string; sourceCommit string; limits object; metrics object; createdAt ISO string.
SNAPSHOT	schemaVersion, complete, boards, jobs, createdAt. Jobs include the full scanned public content before filters. Use a snapshot only when complete=true.
CHANGES	counts of saved new/updated/closed rows and rows with a non-null changeType. Empty in normal mode.

Summary counts are scanned, selected, delivered, new, updated and closed. Each board entry has board, company (or null), status, sourceCount (or null), scanned and error (or null). Metrics include requests, retries, bytes and durationMs. Limits show the active bounds. These records help you tell a valid empty scan from a technical error.

Pricing and limits

The price is $0.027 per start + $0.0036 per saved row ($3.60 per 1,000 rows, plus start fees). A closed job is one saved row too. Role text has no extra fee. Under pay-per-event pricing, the event price covers normal platform usage.

Scenario	Saved rows	Event price
Empty board, no matching title, or no changes	0	$0.027
One job	1	$0.0306
Ten jobs in one run	10	$0.063
Full tested run	53	$0.2178
100 saved rows across at least two runs	100	At least $0.414
1,000 saved rows across at least 19 runs	1,000	At least $4.113

Each run returns at most 53 rows. The 100 and 1,000 examples describe repeated runs, not one larger run. They do not promise that a given company has that many jobs. Retries or fewer rows per run can mean more start fees.

Set an Apify maximum charge when starting a run. If it cannot cover all rows, the SDK stops saving rows within that limit. The scan is partial and its snapshot is not a safe baseline. If a run fails after it starts, the start fee and already saved rows can still be charged. No extra rows are charged for the KVS snapshot.

Use cases

• Keep a job feed up to date: use globalId, title, location and jobUrl.
• Watch hiring changes at companies you know: compare full snapshots and read changeType.
• Find role text for research: keep descriptionText and the source link.
• Send a clean table to a spreadsheet: export Dataset and use jobId to avoid duplicates.

Monitoring

1. Run with incremental: true and no previousRunId. This saves all current jobs as new.
2. Check RUN_SUMMARY.complete and save the run ID only when it is true.
3. Start the next run with the same boards and that ID in previousRunId.
4. Use its ID as the next baseline only if it is complete. Keep the old baseline after an error or cap.

If you use a schedule, your workflow must pass the latest complete run ID to the next run. A fixed scheduled input that keeps the same old ID will keep comparing against that old snapshot. The Actor does not set up a schedule or change your task input for you. Previous runs and their snapshots must still exist.

API and integrations

Send the same input JSON to the Apify API. Use the run's Dataset for job rows and its key-value store for summary and snapshot records. You can export Dataset as JSON, CSV or Excel in Console. Map globalId to your row key and changeType to your update action.

The Actor can be called by an API or an Apify client. In an MCP workflow, use an Apify Actor-run tool that accepts this Actor and its input. This Actor is a batch job; it does not host its own MCP server. No integration or webhook is enabled by a run.

See Apify scheduling for recurring runs and the public HireHive Jobs API reference for source fields.

Limits and partial scans

The tested maximum is four boards and 53 saved rows. Limits are 200 scanned jobs, 16 source requests including retries, 90 seconds of source work, 2 MiB per response and 4 MiB of total response data. A job description above 60,000 characters makes that scan partial. Runs use 256 MB; the recommended platform timeout is 120 seconds. A 300-second Store test or larger finite positive timeout is supported without extending the 90-second source deadline; source requests retry temporary failures at most twice. Larger memory or an unbounded timeout is refused before source work.

This Actor supports standard public tenant.hirehive.com boards. It does not support custom career domains, dashboard APIs or automatic company discovery. It refuses redirects instead of following an unknown host.

• complete: all source pages were checked and all selected rows were saved. Zero rows can be valid.
• partial: some data was read, but a source, output, charge or runtime limit stopped the full result. Keep your last complete baseline.
• not_found: all requested boards returned HTTP 404. No job rows; snapshot unsafe.
• error: invalid input, invalid previous snapshot, no readable source, or a runtime failure. Check the summary; a platform failed run may have incomplete storage.

The source can change while a scan runs. Count checks detect many such changes, but the upstream API has no atomic snapshot token. Two pages can change without changing the total count. Hashes track the content returned in the scan, not a guaranteed point-in-time database copy.

Troubleshooting

Dataset is empty

Check RUN_SUMMARY. Complete with zero delivered rows means no title matched, no change was found, or a valid board had no jobs. not_found means the board did not exist. error means reading or validation failed. Remove filters or check the board URL before retrying.

A previous run is refused

Check that it belongs to this Actor, succeeded, has SNAPSHOT.complete=true and has exactly the same boards. Use your last complete run. Do not use a truncated result.

A board is blocked or changed

A 403 or bad source schema is an error. Open the public board to confirm its format and availability. Retry later or report the run ID. The Actor does not turn an access block into an empty board.

Some jobs are missing

Check OUTPUT_LIMIT, CHARGE_LIMIT, SOURCE_COUNT_MISMATCH or the board error. Increase maxResults within its allowed range or reduce the board scope. Start a new baseline after changing boards.

FAQ

Do I need a HireHive API key?

No. The Actor reads public board content. It does not use the private HireHive dashboard API.

Does an empty scan cost money?

Yes. Every started scan has the start fee. A filter or no-change result still reads the full source.

Does a closed row prove that a job was filled?

No. It means the ID left the public board between two complete checks. The company may have closed or replaced it for another reason.

Can it detect a description change when I hide descriptions?

Yes. includeDescription controls Dataset only. The complete description stays in the snapshot and its hash.

Is it a replacement for every multi-ATS scraper?

No. It supports public HireHive boards and their source fields. It does not add multi-ATS discovery, candidate records, salary estimates or contact enrichment.

Related Actors

• Recruitee Jobs Monitor & Scraper — read companies whose board uses Recruitee.
• Teamtailor Jobs Monitor & Scraper — read public Teamtailor jobs.

Support

Use the Actor's Issues tab for a question or bug. Include the run ID, the board slug and the summary error code. Do not send API tokens, cookies, resumes or private account data.