ai-job-search Puts the Job Hunt in a Forkable Claude Code Repo
A forkable Claude Code framework that evaluates postings, tailors CVs and preps interviews on a local machine — the editable pipeline is worth more than the letters it writes.
“The job search that runs on your machine” is ai-job-search’s whole pitch: a forkable Claude Code framework that evaluates postings, tailors CVs, writes cover letters and preps interviews. The source page lists about three stars total and 1,397 added in the last day.
The input is the most personal document most of us own: employment history, the gaps between jobs, salary expectations, the names of people who will vouch for the claims. Hosted job tools tend to want it uploaded. A fork keeps it in a folder, and the instructions that handle it should be text I can read and cut.
My guess before any of it runs: the cover letters will be ordinary, and the structure will carry the value. Matching a CV to a posting is a comparison task, and comparison is what a coding agent holding both documents does well. Writing a cover letter is generation, in a market already drowning in generated cover letters.
The step I would keep is the evaluation pass: paste a posting, get a blunt read on fit before spending an evening on the application. Forking earns its keep there, because the criteria are mine to edit and every field weights them differently. The editable pipeline is the product; the letters are a by-product.
Skip it if the appeal is software that applies on your behalf, or if reading a prompt before it touches an employment history sounds like homework. The realistic first evening is unglamorous: a cloned folder, one prompt open in an editor, a line half-rewritten. Rewrite one instruction in the CV-tailoring prompt before it goes near a real posting.