Stop Spraying Generic Applications
Sending the same CV to every posting is how applications disappear without a reply. AI Job Search is a free, open-source Claude Code framework that treats each job individually: it scores the posting against your real background, tailors a CV and cover letter for that exact role, has a second Claude review them, and rehearses the interview with you. Its creator built it for their own job search.
Score, Tailor, Then a Second Claude Reviews
Each posting gets a score from 1 to 5 against your actual CV — skills, experience, location and the direction you want to go. One Claude agent then drafts a tailored CV and cover letter, written in LaTeX and output as PDF, and a second Claude reviews both the way a picky recruiter would.
It Won't Invent Experience
The point of the whole pipeline is its anti-fabrication gate: it will not add experience you don't have. Every tailored CV stays a truthful account of your background, just framed for the role in front of you.
A Mock Interview, Run Locally
/interview builds a stage-specific prep pack from the real posting and your materials, maps likely questions to your STAR stories, and runs a mock interview on the hardest ones. Everything runs locally in Claude Code, so your resume never leaves your machine. It's free and MIT-licensed.
The commands follow the search from start to finish: /setup builds your profile, /scrape and /apply find and tailor, /interview rehearses, and /outcome tracks results to calibrate future scoring.
You Still Hit Submit
It scores, drafts and preps, but a person stays in the loop and sends every application. The headline "69 applications, 20 interviews" is the creator's own result — yours will vary.
What's in the Guide
How the scoring and drafter–reviewer pipeline work, what the anti-fabrication gate does, how the mock interview prep is built, the full command list, and a direct link to the repo.
The full field guide is in the PDF.
Get the Guide
Drop your email below and we'll send it straight to your inbox.