LinkedIn JD Extractor

A chrome extension to scrape information from a LinkedIn job post. It's useful for bookkeeping, and theoretically for human-in-the-loop automation. It breaks sometimes due to obfuscation.

Easy install (just use it)

1. Open Chrome and go to chrome://extensions/
2. Turn on Developer mode
3. Click Load unpacked
4. Pick this folder (LinkedIn Extension)
5. Open a LinkedIn job page, then click the extension icon.

Build

Use make build.

It creates:

• build/unpacked/ for Load unpacked
• build/LinkedIn-JD-Extractor-<version>.zip for upload/distribution
• build/version.json with the built artifact metadata

Usage

1. Navigate to any LinkedIn job page (e.g., linkedin.com/jobs/...)
2. Click on a job listing to view its details
3. Click the extension icon in your toolbar
4. View extracted job information in the popup
5. Use Copy to Clipboard or Download JSON to export the data

If LinkedIn changes its markup:

window.__linkedinJDExtractor.debug.testSelectors()
window.__linkedinJDExtractor.extract().then(console.log)
window.__linkedinJDExtractor.debug.findSalary()
window.__linkedinJDExtractor.debug.findDescription()
window.__linkedinJDExtractor.debug.findTitle()
window.__linkedinJDExtractor.debug.findCompany()

1. Open content.js
2. Update the SELECTORS object
3. Reload the extension in chrome://extensions/
4. Test on a LinkedIn job page

Optional Playwright tooling is available if you need an authenticated DOM snapshot:

npm install
npm run auth:linkedin
npm run probe:linkedin -- --url "https://www.linkedin.com/jobs/search/" --wait-ms 8000

This writes output to debug/linkedin-dom/<timestamp>/. Treat .auth/ and debug/linkedin-dom/ as sensitive.

Parser fixture checks

After saving one or more probe snapshots, run the parser against those DOM fixtures:

npm run test:parser -- --fixtures debug/linkedin-dom-test --write-actual

Each fixture may include an expected-export.json sidecar. When present, the runner compares the extracted fields against it and reports diffs. Without an expected sidecar, it prints the parsed result so you can inspect what would export.

Useful flow:

npm run probe:linkedin -- --url "https://www.linkedin.com/jobs/search-results/?currentJobId=4401764934" --wait-ms 8000 --out-dir debug/linkedin-dom-test
npm run test:parser -- --fixtures debug/linkedin-dom-test --write-actual

Notes

• This extension only works on LinkedIn job pages
• Some fields break on some listing paths
• LinkedIn's structure may change; selectors may need updates
• Arc browser users may need to manually trigger extraction after page load