← All runbooks
gooseworks-ai / capabilities-linkedin-job-scraper

LinkedIn Scraper

This runbook finds LinkedIn job postings using the JobSpy Python library and a local `tools/jobspy_scraper.py` wrapper. It turns a requested role, location, company filter, recency window, and output target into a reproducible CSV export with posting metadata and direct URLs. The

agent codexmodel gpt-5.5snapshot python312-uveval programmatic7 stepsv1.0.0

Deploy LinkedIn Scraper to your jetty.io

One-click installs this runbook into a collection on your Jetty account. You can run it from the Spot dashboard, schedule it, or pipe inputs in via the API.

Deploy on jetty.io →View source
The shape of the run

7 steps · start to finish.

  1. 1
    Step 1

    Environment Setup

    ▶

    Install the runtime dependency and create the output directory.

    mkdir -p /app/results
python3.12 -m pip install -U python-jobspy --break-system-packages

    Verify the scraper wrapper is available. If it is missing, copy it from the source skill assets before running.

    test -s tools/jobspy_scraper.py || cp skills/linkedin-job-scraper/scripts/jobspy_scraper.py tools/jobspy_scraper.py
test -s tools/jobspy_scraper.py
  2. 2
    Step 2

    Resolve Search Inputs

    ▶

    Identify the search term, location, result count, recency, company IDs, description requirement, job type, and remote-only flag from the user request. If a request is underspecified, choose pragmatic defaults and record them in `/app/results/summary.md`.

  3. 3
    Step 3

    Construct the Scraper Command

    ▶

    Build the command with only the filters that apply. Always write the CSV to `/app/results/linkedin_jobs.csv` unless the user requested a different file under `/app/results`.

  4. 4
    Step 4

    Run the Scraper

    ▶

    Execute the command and capture the terminal output for the summary. Example:

  5. 5
    Step 5

    Interpret Results

    ▶

    Read the CSV and summarize the outcome in `/app/results/summary.md`:

  6. 6
    Step 6

    Iterate on Errors (max 3 rounds)

    ▶

    If installation, scraper execution, or result validation fails, iterate up to max 3 rounds. After each fix, rerun the failed step and update `/app/results/validation_report.json`.

  7. 7
    Step 7

    Write Validation Report

    ▶

    Write `/app/results/validation_report.json` with this shape: