ai-job-search vs career-ops
A job-application framework built ON Claude Code: fork it, fill in your profile, and the agent evaluates postings, tailors CVs, writes cover letters and preps interviews — locally. — versus — Turns your coding CLI into a job-search command center: scans Greenhouse/Ashby/Lever, scores listings A-F into a 1-5 rubric, tailors ATS-ready CVs, tracks applications. Claude Code, Codex, OpenCode+.
Same job — your coding agent runs the job hunt. ai-job-search is fork-and-fill: a framework you shape; career-ops is the finished product — structured A-F rubric, portal scanners, ATS PDF generation, 17 languages.
| ai-job-search | career-ops | |
|---|---|---|
| Stars | 27k | 62k |
| Forks | 9.0k | 12k |
| Language | TypeScript | JavaScript |
| License | MIT | MIT |
| Last activity | 2 days ago | today |
| Topics | agents | agents, skills |
| Curated connections | 1 | 1 |
ai-job-search — the curator's take
The best example yet of a 'fork-and-own' vertical agent: your data stays on your machine, the workflow is readable Markdown you adapt, and 22k stars say the shape resonates. Even if you never job-hunt, read it as a template for building personal agent workflows on Claude Code. NOT a service — you run and maintain it, and output quality tracks the effort you put into your profile files; garbage in, generic cover letters out.
career-ops — the curator's take
The poster child for consumer-shaped coding-CLI use — the insight that Claude Code is a general agent harness and a job hunt is just a pipeline, validated by WIRED/Business Insider coverage and 62k stars. The rubric design is genuinely good against AI sycophancy: five weighted A-F blocks plus a separate legitimacy check that can never inflate the score. Know what it is NOT: an auto-apply bot — it evaluates, tailors and tracks, deliberately leaving the applying to you; and a tailored CV is only as honest as the profile you feed it. 740 listings → 100 CVs → 1 offer is the author's own funnel, not a promise.