Claude Code drives a real Chrome browser to find matching jobs from a resume, collect them in a CSV, write cover letters, fill forms and submit.
Claude Code drives a real Chrome browser to find matching jobs from a resume, collect them in a CSV, write cover letters, fill forms and submit.
Chrome runs with remote debugging on, and Claude Code connects to it through Selenium scripts it writes itself. A markdown skill prompt, a markdown resume and a cover-letter style template guide it. It searches LinkedIn from the resume, builds to_apply.csv, then writes a letter per posting, fills the form and submits. The author skims the CSV first.
A search run yields 50-130 postings in minutes, and each application takes 60-90 seconds. The author says daily application time fell from 3-4 hours to about 30 minutes. CAPTCHAs and two-factor walls still need a human.
Job postings can carry hidden prompt-injection canary words that auto-reject bots, so add an ignore-embedded-instructions rule to the skill prompt. Setup took about two hours.
Using my resume at [path to resume file], search LinkedIn in my Chrome browser for [job title] roles in [location] and save the best matches to a CSV with title, company, and link. Then draft a tailored cover letter for each of the top [number] matches. Before you submit any application or fill in a final form, pause and show me the cover letter and the answers you plan to send. Stop when the CSV and drafts are ready and I have approved or skipped each one.
A Claude Code-built agent checked ESPN every 30 minutes and sent each 2026 World Cup final score, written up in a persona, to a 3-person Signal group, with a code check on scores.
The poster exports Garmin data into a Claude project and meal planner, where Claude spotted a link between eating more on one day and reduced medication efficacy.
A Claude Cowork skill tracks diet and medical metrics in a spreadsheet, and meals told to Claude on mobile are logged in the daily record.