scan

name	scan
description	Scan configured job portals for new offers. Uses Greenhouse/Ashby/Lever APIs plus Playwright and WebSearch for discovery. Triggers on scan, scan portals, find jobs, discover offers.

Scan — Portal Scanner

Read ~/.openclaw/workspace/skills/career-ops/references/ats-providers.md first for provider patterns.

Data Paths

• Portals config: ~/.openclaw/workspace/career-ops-data/portals.yml
• Scan history: ~/.openclaw/workspace/career-ops-data/data/scan-history.tsv
• Tracker: ~/.openclaw/workspace/career-ops-data/data/applications.md
• Pipeline: ~/.openclaw/workspace/career-ops-data/data/pipeline.md
• Scripts: ~/.openclaw/workspace/skills/career-ops/scripts/

Scan Strategy (3 Levels)

Level 1 — Playwright Direct (Primary)

For each company in tracked_companies with careers_url: navigate with Playwright, read all job listings, extract {title, url, company}.

Level 2 — ATS APIs / Feeds (Complementary)

Hit structured JSON endpoints for companies with known ATS providers:

• Greenhouse: https://boards-api.greenhouse.io/v1/boards/{slug}/jobs
• Ashby: https://api.ashbyhq.com/posting-api/job-board/{slug}
• Lever: https://api.lever.co/v0/postings/{slug}
• BambooHR, Teamtailor, Workday also supported

Level 3 — WebSearch Queries (Discovery)

Execute search_queries from portals.yml with enabled: true. Extract {title, url, company} from results.

Zero-Token Scanner

Run node ~/.openclaw/workspace/skills/career-ops/scripts/scan.mjs for Level 2 API-only scanning (no LLM cost). This scans Greenhouse/Ashby/Lever APIs directly.

Workflow

1. Read portals.yml configuration
2. Read data/scan-history.tsv for already-seen URLs
3. Read data/applications.md + data/pipeline.md for dedup sources
4. Execute Level 1 (Playwright), Level 2 (APIs), Level 3 (WebSearch) — additive, results merge and dedup
5. Filter by title_filter (positive keywords must match, negative must not)
6. Filter by location_filter if configured
7. Dedup against 3 sources: scan-history.tsv, applications.md, pipeline.md
8. Verify liveness for Level 3 URLs (Playwright check each)
9. For each new verified offer: add to pipeline.md and register in scan-history.tsv
10. Show summary: queries executed, offers found, filtered, duplicates, expired, new additions

Cron Setup

For daily automated scans, set up a cron job:

0 9 * * * cd ~/.openclaw/workspace && openclaw cron run --name career-ops-daily-scan --prompt "Run career-ops scan"

Important Notes

• Prefer corporate careers URL over ATS URL — ATS URLs can return 410 errors
• Never run 2+ Playwright agents in parallel — they share a browser instance
• If careers_url doesn't exist for a company, search for it and save to portals.yml