A job search tracker for one person is a board of jobs, contacts and notes next to a resume editor that exports a PDF. Add an LLM call that reads a pasted job description, scores your resume against it and rewrites bullet points from facts already in the resume, and you have Huntr's paid core in one sitting. The loose ends are the browser clipper, which a bookmarklet covers for the basics, autofilling application forms, and job insights. Several open-source trackers and resume builders exist, so you may only need to wire one of them to your own model.
Before you buildCheck Huntr Free first: it tracks up to 100 jobs and gives 2 job-tailored resumes, which may cover a short search.
Build me a personal job search tracker to replace Huntr. Requirements: - Node.js 22 + TypeScript: a Fastify server and a React board built with Vite. `npm start` serves http://localhost:4800. - Storage: better-sqlite3 in ./data/jobs.db for jobs (title, company, URL, location, salary text, posting text, status, source, dates), contacts per job, notes, a stage history, tags and documents (resume and cover letter versions). Files go to ./data/files. - Board: drag-and-drop columns Wishlist, Applied, Interview, Offer and Rejected (`@dnd-kit`), a filterable table view, and a follow-up date per job with an "overdue" list. - Capture: a bookmarklet that posts the page title, URL and selected text to POST /api/clip (token in .env); without a selection the server fetches the page text with `@mozilla/readability`. Dedupe by URL. - Resumes: the master resume is a JSON Resume file at ./data/resume.json; render it to a one-page PDF with Playwright from two plain, single-column HTML templates with real selectable text. - AI: any OpenAI-compatible endpoint (LLM_BASE_URL, LLM_API_KEY, LLM_MODEL in .env). Per job: a match score 0-100 with missing keywords, a tailored resume that only rewords what is already in the master resume (never invents jobs, skills or numbers) shown as a diff, and a cover letter draft. Keep every generation with the prompt that produced it. - Metrics: applications per week, response rate and days per stage from the stage history. - Export: all jobs as CSV and the whole database as one JSON file. - Single user. No accounts, no telemetry, local-first: only the model calls leave the machine. - Out of scope: autofilling application forms, job board search, coach accounts, calendar sync. - README: setup, the bookmarklet, the .env keys and where the data lives.
Opens your agent with the prompt filled in, you press enter. Or copy it and paste it into an empty folder.
The first prompt builds the core. Run these one at a time in the same folder to grow it into your own Huntr.
Add a Huntr importer to the job tracker in this folder. Requirements: - `jobs import-huntr <zip>` reads the Download My Data ZIP, finds the Job, Contact and Activity CSV files by header names and loads them into SQLite. - Map Huntr list names to the board columns, keep created dates, and link each contact to its job. - Show a summary of new, duplicate (same URL or same company and title) and unmapped rows, ask to confirm, and make a second run change nothing. - Update the README with the Huntr export steps.
Add interview prep to the job tracker in this folder. Requirements: - A button on a job that sends the posting and my master resume to the configured model and returns 8 likely questions with a short answer outline for each, using only facts from the resume. - A reusable story bank: short STAR notes I write once and tag by skill; suggest the best 3 per question by tag match, with no model call. - Save everything on the job; export a job's prep as Markdown. - Update the README.
Add reminders to the job tracker in this folder. Requirements: - Every morning a cron job lists jobs in Applied with no stage change for 7 days and any follow-up dates due today, and emails the list through SMTP settings in .env. - A draft follow-up message per job, written by the configured model from my notes, saved on the job and never sent automatically. - Snooze a reminder for 3 or 7 days from the email link, signed with an HMAC token. - Update the README.
A job hunt is short and stressful, and Huntr keeps the board, tailored resumes and cover letters in one place, with a clipper that saves a posting in one click. The Pro price is small next to a salary, and you can drop back to Free afterwards without losing data.
Estimate · Runs on your machine; model tokens for tailoring and cover letters are the only running cost. The build is token usage on your agent plan.
| Plan | Monthly | Annual, per month | What you get |
|---|---|---|---|
| Free | $0 | $0 | Unlimited base resumes with PDF export and all templates, 2 job-tailored resumes and 2 application packets, basic matching and scoring, a tracker for up to 100 jobs, Chrome clipper, contacts, application autofill, 100 document uploads. |
| Pro | $40 | — | Everything in Free plus unlimited job-tailored resumes, application packets and cover letters, unlimited AI generations, reviews and tailoring, advanced matching and scoring, unlimited job tracking and uploads, advanced job insights. |
Free covers unlimited base resumes, a tracker for up to 100 jobs and 2 job-tailored resumes.
Yes. An AI coding agent such as Claude Code, Codex or Cursor can build a usable personal version in one session with the prompt on this page, running on your own machine.
About $40 a month on the Pro (monthly) plan, $480 a year (checked 2026-10-19).
The Chrome extension that saves jobs from boards and autofills application forms; Advanced job matching and resume scoring checks such as spelling and metrics; Job insights and search metrics charts; Polished resume templates that hold up in applicant tracking systems; Sharing a board with a coach or a career center. If any of those matter to you, keep paying.
Yes: Reactive Resume (open source), Resume Matcher (open source), JobSync (open source).