career-ops

An open-source, local-first AI job search that helps you apply better to fewer roles

career-ops is an open-source AI job-search system that runs locally inside an AI coding CLI such as Claude Code, Codex, OpenCode, or Antigravity. It scans job portals, evaluates each listing into a structured report with a global 1–5 score, tailors your CV, and tracks your applications. This is my fork of career-ops-hq/career-ops, kept as a study in how to package a real workflow as agent skills.

The concepts

Companies use AI to filter candidates; this project gives the candidate a filter of their own. The README sums up its practice in a handful of short principles: send fewer, better applications; prefer signal to volume and evidence to keywords; keep a human in charge; keep the data local.

Two of those are design decisions I’d point to in any agent project:

  • A human decides. The tool tells you whether a job is still open and whether it fits, tailors the CV, and drafts your answers. You press Submit. It does not mass-apply on your behalf.
  • Local-first. Your CV and your application history stay on your machine, inside the CLI you already use.

A “do not apply” result is treated as a good outcome: it gives you the evening back.

How to use it

npx @santifer/career-ops init

Then paste a job you were about to apply to and read the evaluation. See the upstream README and career-ops.org for the supported CLIs and the full workflow. MIT licensed.

Language: JavaScript
Source code: github.com/avnit/career-ops
Upstream: github.com/career-ops-hq/career-ops

Share: X (Twitter) Facebook LinkedIn