Apify Job Boards

Job boards to one table

Disclosure: the author of this skill owns both Actors it routes to (flash_scraper/multi-jobboard-scraper and flash_scraper/remote-job-aggregator). They are pay-per-result Actors on the Apify Store; no referral or tracking parameters are used anywhere in this skill.

Turn "I need job postings" into one validated, deduplicated dataset by routing the request to the right multi-board Actor, sizing the run so the user knows the cost before it starts, checking the returned locations and vacancy identities, and returning the rows with the board each job was found on. Treat the Actor output as raw input to these checks, not as proof that its location filtering or deduplication succeeded.

Example prompts

Prompts this skill handles:

• "Scrape software engineer jobs in Austin from LinkedIn, Indeed and Glassdoor into one spreadsheet, no duplicates."
• "Give me remote Python developer jobs posted this week across the remote job boards, with salary where listed."
• "Watch Stripe's and OpenAI's careers pages and only tell me about new roles."

For the Austin request, try all three requested boards, then validate each returned location and vacancy identity before delivery. If Glassdoor returns only wrong-city or unverifiable rows, deliver the qualified LinkedIn and Indeed rows as a partial result and report Glassdoor's raw, qualified and deduplicated counts as a board failure. Do not claim three-board coverage merely because Glassdoor returned rows.

Out of scope (the boundary):

• "Apply to these jobs for me" or anything that needs a logged-in account, CAPTCHA solving, or personal data of applicants. This skill only reads public postings; for candidate profiles send the user to the LinkedIn workflows in apify/agent-skills ultimate-scraper.

Prerequisites

• Apify account (sign up)
• Authentication via one of:
  • apify login (OAuth, if using the Apify CLI)
  • APIFY_TOKEN environment variable
  • Token from Apify Console → Settings → Integrations

Never paste a token into a URL or into a file inside this skill; pass it as Authorization: Bearer (the CLI does this for you).

Workflow

Copy this checklist and track progress:

Task Progress:
- [ ] Step 1: Get the four anchors (what, where, remote-only?, how many)
- [ ] Step 2: Route to the Actor
- [ ] Step 3: Build the input and state the cost
- [ ] Step 4: Run and wait
- [ ] Step 5: Validate and deliver: locality, vacancy identity, per-board counts and failures

Step 1: Get the four anchors

Ask these as one block; do not start a run without them.

1. What — the role or keyword(s), e.g. data analyst, registered nurse. Several are fine.
2. Where — a city/region string like Chicago, IL, or "remote".
3. Remote-only? — yes/no. "Yes" with no city routes to the remote aggregator (Step 2).
4. How many — a per-board cap. Default to 20 per board for a first run; ask before going above 100.

Optional follow-ups, only if the user raises them: posted-within window, salary required, job type, exclude staffing agencies, company watch list, Slack/Discord webhook for alerts.

Step 2: Route to the Actor

Rule of thumb: a city or country in the request → multi-board scraper. "Remote" and nothing else → remote aggregator. Both Actors run without an API key, login or cookies.

Check the live input schema before building input (fields change; the schema wins over this file):

apify actors info "flash_scraper/multi-jobboard-scraper" --input --json \
  --user-agent apify-awesome-skills/apify-job-boards 2>/dev/null

apify actors info "flash_scraper/remote-job-aggregator" --input --json \
  --user-agent apify-awesome-skills/apify-job-boards 2>/dev/null

Step 3: Build the input and state the cost

Multi-board (city/region search). Field names as in the live schema:

{
  "searchTerm": "data analyst",
  "location": "Chicago, IL",
  "sites": ["linkedin", "indeed", "glassdoor", "muse"],
  "maxResults": 20,
  "countryIndeed": "usa",
  "strictKeywordMatch": true
}

• sites accepts: linkedin, indeed, glassdoor, muse, remotive, jobicy, himalayas, hn_hiring, devitjobs_us, devitjobs_uk, remoteok, weworkremotely, working_nomads. Leave out google, zip_recruiter, bayt, bdjobs, naukri: they are in the enum for backwards compatibility but blocked at the source, and the run report will say so.
• maxResults is per board, so 4 boards × 20 = up to 80 rows before deduplication.
• strictKeywordMatch: true drops rows whose title and description never mention the search term (LinkedIn in particular returns loosely related postings). Filtered rows are not billed.
• isRemote: true restricts to remote postings and auto-adds the remote boards.
• Several roles or cities: use searchTerms / locations arrays instead of the singular fields.
• Company watch: "atsCompanies": ["stripe", "openai"] reads their Greenhouse/Lever/Ashby boards directly; add "onlyNewJobs": true on a schedule so each run delivers only postings it has not sent before (the Actor remembers what it delivered for 90 days).
• Alerts: "webhookUrl": "https://hooks.slack.com/..." posts a digest of new rows to Slack, Discord or any webhook.

Remote aggregator. Field names as in the live schema:

{
  "searchTerms": ["python"],
  "boards": ["remoteok", "weworkremotely", "remotive", "jobicy", "himalayas", "hn_hiring"],
  "maxItems": 100,
  "matchDescriptions": true,
  "postedWithinDays": 7
}

• matchDescriptions: true matches the keyword in descriptions as well as titles (the README's measured example: 28 rows title-only vs 194 with descriptions for python).
• salaryMinAnnual, seniority, countries, excludeKeywords are cheap filters that run before billing.
• onlyNewJobs and webhookUrl work the same way as above.

Cost, stated before the run. Both Actors bill per delivered row; filtered and deduplicated rows are not billed. Read the current price from the Store Pricing tab (the schema fetch in Step 2 also returns the pricing block). At the time of writing the free-plan rate is $0.005 per job on the multi-board scraper and $0.002 per job on the remote aggregator; paid plans pay less. So a first multi-board run of 4 boards × 20 rows costs at most about $0.40 and usually less after deduplication; a 100-row remote run about $0.20. If the user asks for more than 500 rows, say the number and confirm before running.

Step 4: Run and wait

apify actors call "flash_scraper/multi-jobboard-scraper" -i '{"searchTerm":"data analyst","location":"Chicago, IL","sites":["linkedin","indeed","glassdoor","muse"],"maxResults":20,"strictKeywordMatch":true}' \
  --json \
  --user-agent apify-awesome-skills/apify-job-boards \
  2>/dev/null

apify actors call "flash_scraper/remote-job-aggregator" -i '{"searchTerms":["python"],"maxItems":100,"matchDescriptions":true,"postedWithinDays":7}' \
  --json \
  --user-agent apify-awesome-skills/apify-job-boards \
  2>/dev/null

Typical durations: a 4-board, 20-per-board multi-board run finishes in about a minute; the remote aggregator answers a 100-row query in under a minute, and a 20-row query on the four single-request boards in a few seconds. The JSON output contains defaultDatasetId; fetch the rows with:

apify datasets get-items DATASET_ID --format json \
  --user-agent apify-awesome-skills/apify-job-boards 2>/dev/null