Agent Upstart

Cool things people are building with agents — and practical help getting the most out of them.

Skills · last verified 2026-10-24

career-ops-hq/career-ops — score a posting before you apply, with Claude Code

github.com/career-ops-hq/career-ops · open-source AI job search agent · fit scoring, tailored CVs, portal scanning, application tracking

What it does

Career-ops is a job search agent that runs inside your AI coding CLI. You paste a job posting, and it checks whether the posting is still open, scores the role against your real CV on a 1–5 scale, and drafts a tailored CV and cover letter. You read them and you press submit. The README states the design plainly: it never auto-submits an application, never sends an email, and never uploads your data anywhere — every application stays on your machine, and your CV goes only to the AI provider you chose.

The author's own account of his search, stated in the README, is 740 listings evaluated, 68 applied, 12 interviews, and 1 offer. That is one person's result, not a benchmark and not a promise about yours.

The score is the decision tool. A posting that scores below 4.0 out of 5 gets a do-not-apply recommendation, which you can override; the README is explicit that it will not push you to apply below 4.0. The evaluation behind the score covers the role summary, the CV match with how much each requirement matters for that posting, a level strategy, compensation research, personalization, interview preparation, and a posting-legitimacy check (whether it is still open, whether it is a repost) plus a work-authorization signal that flags an explicit no-sponsorship posting as a hard blocker.

Beyond single postings, a scan mode searches 100+ pre-configured companies and major job boards for new openings, and a dashboard view lets you browse the pipeline you have built. Interview preparation, offer reading, follow-up tracking, and rejection-pattern analysis are included as further modes, all driven from the same command router.

Install

Prerequisites: Node.js, which ships the npx runner used below. The fastest route, from the repo README, is one command:

npx @santifer/career-ops init

This clones the latest release into a career-ops folder and installs dependencies. Then open your AI coding CLI in that folder:

cd career-ops
claude

Claude Code is the default, but the README lists Codex, OpenCode, Antigravity, Grok, Qwen, and others as supported. On first launch it walks you through setup — your CV, your profile, and your target roles — just by chatting, with nothing to edit by hand. A manual route also exists: clone the repo, run npm install, and install Chromium for PDF generation with npx playwright install chromium.

With setup done, the first run is a single paste. In a CLI that registers slash commands:

/career-ops https://company.example/jobs/123

Or paste the posting text directly. The auto-pipeline evaluates it, writes a report, generates the tailored CV PDF, and adds a tracker entry. Other commands from the same router: /career-ops scan to discover new openings, /career-ops cover for a standalone cover letter, /career-ops tracker for an overview of your applications, and /career-ops interview-prep for company-specific preparation.

One CLI difference matters. In Codex, slash commands are not guaranteed, so the README says to ask for the mode in plain language instead — for example, "Run the career-ops scan mode" or "Evaluate this JD with career-ops auto-pipeline" followed by the posting. The same applies to any CLI where the slash command does not appear.

Two limits to know before you start. The README warns that the tailored CV must be read before sending: a code check blocks a CV whose numbers or facts appear in neither your CV nor your proof points, but it cannot judge every rewording yet. And the fit score is a recommendation, not a verdict — the README's own disclaimer says evaluations are recommendations, not truth.

Verified against github.com/career-ops-hq/career-ops (README: Your turn, What career-ops does for you, What career-ops will not do, Features, Quick Start, Usage, Codex Integration, FAQ, Disclaimer), 2026-10-24.

Trust score: not scored

No score is given for this page. The repository holds many mode files and scripts, and those files were not fully read in this pass, so a scored review would not be honest. What was checked is listed under Verification below.

Verification

Checked 2026-10-24 against the author's own repository. No install was run and no account was connected for this check.

What was checked: Author README (Your turn, What career-ops does for you, What career-ops will not do, Features, Quick Start, Usage, Codex Integration, FAQ, Disclaimer) was read from a fresh clone of the repo. Install commands and command names on this page are copied from the README.

Limits: Install commands checked against the author README; not executed in this pass. The mode files and scripts behind the commands were not read in full, so no trust score is given. The creator's outcome numbers are the author's own account from the README, not an independent measurement. The README's own install and clone commands still name the project's previous path; the current path used on this page is the one the clone was taken from.

Sources: github.com/career-ops-hq/career-ops (README)

Also in this series

Related playbook: Keep your job-hunt repo private

Back to: Skills index