Browse / Job hunt / JobFinderOS
JobFinderOS
matthewprice
Open-source AI job search agents for Claude Code.
Open matthewprice/JobFinderOS → Checked against the four rules.
agentic-aiai-agentsanthropicashbyatsautomationcareerclaudeclaude-codegreenhousejob-boardjob-hunting
Works with any agent
ships AGENTS.md, the cross-tool instruction file.
In its own words
Polls Greenhouse, Ashby and Lever job boards directly, scores roles against your profile, preps interviews, drafts outreach, and writes it all to an Obsidian vault. A recruiter, a crawler and a market analyst, benchmarked against plain LLM crawling.
Find your dream job with Claude Code agents. No coffee breaks, no doomscrolling, no sleep till you're hired!
This project capitalizes on one simple fact: a career search is a numbers game. More real contacts lead to more listings you hear about early, more of those become applications with a person attached, and more of those become interviews. A single human running that chain alone has limited hours in the week. Agents do not. JobFinderOS is a Claude code project that puts agents to work on your search, profiling markets, following companies, crawling job postings, identifying new strategic contacts and preparing you for every interview. Each stage of the job search funnel gets more nurturing than you could feed it yourself, and the judgment stays with you.
This manual covers four things:
- What it is
- Getting up and running
- Using it
- The methodology: how agentic job search works here
Plus a short reference and a note on contributing at the end.
1. What it is
JobFinderOS is a set of agent personas and skills that run inside Claude Code and work for you to identify career opportunities based on your criteria. They watch the market, find roles, keep your pipeline up to date and honest, and get you ready for every interview. They write everything to a folder of Markdown notes, which you can open in Obsidian or any text editor.
Three agents, each with one job.
Coach, the recruiter · Scout, the crawler · Mark, the market analyst
The repository's own description, then 10% of its README. Read the rest on GitHub ↗
Read from the repository's git tree on 2026-08-17.