Thin wrapper around jobspy.scrape_jobs() for generating a daily list of fresh job-posting leads across configured tech verticals and locations. The posting's company is the lead — a company actively hiring is a prospect for your recruitment services.

• daily_leads.py — runner script
• leads_config.yaml — edit this to change verticals, locations, sites, etc.

From the repo root:

pip install -e .
pip install pyyaml
python examples/daily_leads.py

Outputs:

• leads/leads_YYYY-MM-DD.csv — today's new leads
• leads/seen_jobs.csv — running history of every job URL ever emitted; used to dedupe subsequent runs

Run it a second time immediately — the summary should report 0 new leads, confirming dedupe works.

Linux/macOS cron (06:00 local):

0 6 * * *  cd /path/to/JobSpy && /usr/bin/python examples/daily_leads.py >> leads/run.log 2>&1

Windows: use Task Scheduler with the same command.

• Verticals / locations: edit leads_config.yaml. Start small (1-2 verticals x 1-2 locations) to sanity-check, then expand.
• results_wanted: per board, per query. 50 is a good starting point.
• hours_old: keep at 24 for a daily schedule. Increase if you run less often. Note LinkedIn/Indeed restrict combining hours_old with some other filters — the script avoids those combinations by default.
• Proxies: if you hit rate limits, add a proxies key to the config and wire it through to scrape_jobs(proxies=...) in daily_leads.py.

The CSV leads with recruitment-relevant fields: site, company, title, location, job_url, company_url, date_posted, job_type, is_remote, min_amount, max_amount, currency, interval, description, search_term, search_location. All other JobSpy fields are retained after these.