AI Skill MartBrowse skillsPromptsProgramsLearnWhat is a Skill?
 
← Back to list

apply — the job application loop

Drives the jev CLI application loop: fills each queued job form in its own Chrome window, verifies every value on the page, and submits only the forms that are truly ready.

AutomationIntermediate★ 336⑂ 62AI score 9/10Last updated: Oct 19, 2026

What it does

Say /apply or "start applying" and Claude runs the full jev application loop for you.

  • Pre-flight: doctor for environment, status for the queue, /discover if the queue is empty or stale, and it always asks how many to send before submitting
  • Filling: fields are mapped automatically, anything left open is answered by a headless Claude call, and multi-page forms are walked page by page
  • Verification: every value is read back from the rendered page before anything is sent
  • Outcome playbook: a table covering READY/submitted, needs your code, robot checks, not confirmed, filled, not ready, blocked, Workday sign-in handoffs and more
  • Hard limits: never signs in, never touches a password/token/emailed code, never passes a CAPTCHA, never signs an agreement for you
  • Wrap-up: counts sent, what was left to you and why, forms waiting on resume, unconfirmed runs needing reconcile, run cost, and where the applications/*.csv records live

Who it's for

  • Job seekers applying at volume on Greenhouse, Lever, Ashby and Workday
  • People who want a clean CSV audit trail of every application
  • Terminal-comfortable users who want automation with explicit human-in-the-loop boundaries

Examples

  1. Rehearsal on a new machine: "test five forms without sending" → apply --dry --count 5 prints each label and the value shown, so you can fix data/profile.json before anything goes out.
  2. A real run: "apply to my top 10 and submit" → apply --count 10 --submit, then a report of sent / waiting / manual jobs.
  3. A posting you found: "apply to this Lever link" → apply <link> --submit rates it, queues it at the top and fills it.
  4. Fixing a bad answer: edit answers in the profile, re-check with apply --dry <id>, and clear stale reuse with memory --forget.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/qbeka/jev-job-search/HEAD/.claude/skills/apply/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on AI Skill Mart.
Copy the .claude/skills/apply folder from the GitHub repo qbeka/jev-job-search into my ~/.claude/skills/apply/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/qbeka/jev-job-search.git && cd jev-job-search && npm install && mkdir -p ~/.claude/skills && cp -r .claude/skills/apply ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Make sure Node.js (LTS), Git and Chrome are installed.
  2. Clone the repository: git clone https://github.com/qbeka/jev-job-search.git
  3. Install dependencies: cd jev-job-search && npm install
  4. Optional global install of the skill: mkdir -p ~/.claude/skills && cp -r .claude/skills/apply ~/.claude/skills/ (running Claude Code inside the repo already picks it up).
  5. Open Claude Code at the repo root and run /setup to build data/profile.json (experience, education, work authorization, standing answers).
  6. Run npx jev doctor to validate the environment, then npx jev discover to build the job queue.
  7. Always rehearse first with npx jev apply --dry --count 5; once the filled values look right, run /apply for a real submitting run.
View source on GitHub ↗License: MIT