AI Job Search is an open-source, MIT-licensed framework that runs entirely on your own machine via Claude Code. It evaluates job postings against your background, tailors your CV and cover letter for each application, and helps you prep for interviews — all without sending your data to a third-party service. Over 31,000 GitHub stars; fork it and customize the prompts and workflow to fit your own job search.
AI Job Search is an open source framework that runs inside Claude Code on your own machine rather than as a hosted service. You give it your background, point it at job postings, and it evaluates how well each role matches you, tailors a CV and cover letter for the applications worth making, and helps you prepare for interviews. Because it executes locally through an agent you already run, your resume and work history stay on your machine instead of being uploaded to a third party job platform. It is distributed under the MIT license, so the prompts and the workflow are yours to read and rewrite. In practice that is the point. A job search is personal, and the shipped instructions are a starting position you fork to match the roles you actually want. Expect to configure it before it earns its keep, the way you would with any agent framework rather than a finished consumer app.
This fits the middle of an active search, when you are sending enough applications that per role tailoring has become the bottleneck. Rewriting the same CV twenty times by hand is exactly the kind of repetitive judgment work an agent handles well, and running it locally means you can keep the salary history, the reasons you left, and the drafts you would never publish in a file the agent reads and nobody else does. It also suits anyone who dislikes handing a full career history to a job board in exchange for matching. The prerequisites are real: you need Claude Code installed and working, and you need to be willing to edit prompts when the output reads generic. Someone who wants a click and forget service will not enjoy it.
Claude Task Master solves the same class of problem in a different domain, turning a brief into a dependency aware task list so an agent keeps working through a long project instead of drifting. If your search feels disorganized rather than under tailored, that structure may help more than the document generation here. OSINT Skill points the agent outward at researching people and organizations, which is the natural companion when you want to understand a hiring manager or a company before an interview rather than produce application material. Marketing Skills is another skill library aimed at persuasive writing, useful if the weak part of your applications is the pitch itself. AI Job Search is the only one of the four built end to end around the application loop.
Compare with: Claude Task Master, OSINT Skill, Marketing Skills
The catalog records this as free, and the project is MIT licensed, so there is no vendor and no plan. Your real running cost is the agent it sits on top of: you are spending Claude Code usage every time it reads a posting and drafts a document, and a heavy application week is not a trivial amount of that. Budget for the assistant, not for the framework. Cloning it, reading the prompts, and deciding it is not for you costs nothing.
No notes have been added for this tool yet.
No prompting tips have been added yet.
No reviews yet. Be the first!