How to use the Career-Ops Job Search Agent skill

What is an agent skill and what does career-ops do?

An agent skill is a folder containing a SKILL.md file that tells an AI assistant when and how to perform a task. The skill is loaded when your request matches its description, so you do not need to memorise commands.

The career-ops skill is a router. It reads your request, works out which mode applies (evaluating an offer, generating a CV, scanning portals, tracking applications, interview preparation or outreach drafting) and then runs that mode. The skill file itself contains only this routing; the modes, scripts and templates live in the separate career-ops repository.

• Evaluate job posts pasted as a URL or as a job description
• Scan job portals for new matches
• Generate tailored CVs and PDFs
• Track applications
• Prepare for interviews and debrief afterwards
• Draft outreach emails

Install the career-ops skill

Download the skill ZIP from this page and unzip it. Then place the resulting career-ops folder in one of these locations: ~/.claude/skills/ to make it available across your projects, or .claude/skills/ inside a specific project folder to limit it to that project.

If you use the Claude apps instead, open Settings, go to Capabilities, then Skills, and upload the ZIP there.

Installing the skill alone is not enough. The skill needs the career-ops repository installed locally before any mode can run. If the agent finds no career-ops install, it will offer to set one up with the command below. It needs Node.js and downloads software, so it asks before running anything.

• Requires Node.js for the career-ops install and its scripts
• Install the repository with npx @santifer/career-ops init, or by git clone of the career-ops repository
• Prepare a CV and profile file, which you supply yourself

Trigger the skill with natural requests

You do not need a special command. Describe what you want in plain language, and the agent picks the matching mode. You can also name a mode directly, for example by asking Codex to run the career-ops scan mode. In CLIs that register slash commands, the router can be invoked as /career-ops.

Paste a job URL or the full job description and ask for an evaluation. Other requests that trigger the skill include asking to scan portals, to generate a CV or PDF, to track applications, to prepare for an interview, or to draft an email.

• Evaluate this JD: [paste the job description]
• Evaluate this job with career-ops: https://company.com/jobs/123
• Run the career-ops scan mode and summarise new matches
• Run the career-ops pdf mode for the latest evaluated role
• Run the career-ops tracker mode and summarise the current statuses

What you need before it works

Node.js is required because the career-ops install and its scripts run on it. The repository must be installed locally. You also need a CV and profile file, which you provide. The skill works with AI CLIs such as Claude Code, Cursor, Pi or Codex.

Your CV and personal data are stored as local files in a user data folder on your machine. The career-ops tool is not a hosted service.

• Node.js
• The career-ops repository installed locally
• A CV and profile file you supply
• An AI CLI such as Claude Code, Cursor, Pi or Codex

Safety notes: scripts and network access

This skill runs scripts and uses the network. Scanning portals and any automated contact can breach a site's terms, or misrepresent you to employers, so review anything before it is sent. Treat outreach drafts as drafts to check, not messages to send unseen.

Keep your personal CV data in the local user data folder and do not share that folder. Be aware that the installer downloads the latest release from the career-ops GitHub repository and installs its dependencies.

• Review outreach emails and applications before they leave your computer
• Check the terms of any job portal before asking the agent to scan it
• Only run the install command from a source you trust

Tips for better results

Give the agent complete context: paste the full job description rather than a summary, and make sure your CV file is up to date before generating a PDF. For repeated work, open your AI CLI inside the career-ops folder so the setup is already in place.

Ask for one mode at a time when you are learning, then move to chained requests such as evaluating a role and then generating a tailored CV for it.

• Paste the full job description for a more accurate evaluation
• Keep your CV and profile file current
• Open the AI CLI inside the career-ops folder for daily use