skill
claude-code-job-search
A Claude Code skill that finds remote software engineering jobs, filters against your profile, and drives an apply/reject triage workflow.
Set up
Skills are installed by placing their folder where your agent loads skills.
The download comes from github.com/Yodablues/claude-code-job-search itself, not from our server, so what you get is exactly what its author published.
Install instructions come from the README
README
synced Oct 25, 2026claude-code-job-search
A Claude Code skill that finds remote software engineering jobs, filters them against your resume and salary requirements, and walks you through a triage workflow — all from your terminal.
It queries public job APIs (RemoteOK, Remotive, Greenhouse, Ashby, Lever), applies a coarse keyword filter, then hands the results to Claude to judge real fit against your profile. You get a ranked shortlist and go through it one-by-one: apply or reject.
Skills included
/job-search — Find and triage jobs
- Automated sourcing from 200+ company ATS feeds + RemoteOK + Remotive, parallelized
- Coarse pre-filter by seniority, role type, tech stack, and remote status
- Claude-powered fit scoring against your actual profile (not just keyword matching)
- Already-applied detection — skips companies you've already applied to (via tailored resume files on disk or a manual exclusion list)
/resume-cover — Tailored resume & cover letter
- Fetches the job posting and extracts requirements automatically
- Reads your master resume as the single source of truth
- Tailors the resume — rewrites summary, reorders bullets, mirrors the JD's language
- Writes a cover letter with bold lead-in paragraphs mapping your experience to their requirements
- Outputs .docx files with clean Calibri formatting, ready to submit
- Never fabricates experience — addresses gaps honestly
Installation
# Clone into your Claude Code skills directory
git clone https://github.com/Yodablues/claude-code-job-search ~/.claude/skills/job-search
Job search setup
Edit the CONFIG section at the top of find_jobs.py:
SENIORITY,ROLE,TECH— coarse keyword filters for titles and descriptionsGREENHOUSE,ASHBY,LEVER— company ATS tokens (200+ included, add more freely — unknown tokens are skipped)MANUAL_APPLIED— set of company name tokens to exclude (already applied)APPLIED_DIR— directory to scan for tailored resume/cover letter files (auto-exclusion)
Resume & cover letter setup
cd ~/.claude/skills/job-search/resume-cover
cp config.example.json config.json
Edit config.json with your details:
{
"name": "Your Name",
"location": "City, State",
"email": "you@example.com",
"phone": "555-555-5555",
"linkedin": "linkedin.com/in/your-profile",
"website": "yoursite.com",
"github": "github.com/you",
"source_resume": "/path/to/Your_Resume.docx",
"output_dir": "/path/to/output/directory"
}
Usage
In Claude Code, just say:
find me jobs
Or any variation: "go find jobs", "any new roles", "search for jobs".
Claude will:
- Run the gather script to pull fresh postings
- Read the results and aggressively filter against your profile
- Present a ranked shortlist
- Walk through each match for you to apply or reject
When you say "apply" on a role:
/resume-cover <job-posting-url>
Claude will fetch the JD, tailor your resume and cover letter, and output .docx files.
Requirements
- Claude Code (CLI, desktop app, or IDE extension)
- Python 3.8+ (stdlib only for job search — no pip install needed)
- python-docx (for resume/cover letter generation:
pip install python-docx)
Adding company ATS feeds
The script queries company career pages via their public ATS APIs. To add companies, put their board token/name in the appropriate list in find_jobs.py:
- Greenhouse: The URL slug from
boards.greenhouse.io/{token}— e.g."stripe","discord" - Ashby: The company name from their job board — e.g.
"Linear","Vercel" - Lever: The URL slug from
jobs.lever.co/{token}— e.g."netflix"
Unknown or invalid tokens are silently skipped, so it's safe to add speculative entries. 200+ companies are included out of the box.
How it works
"find me jobs"
│
├── find_jobs.py # GATHER: hits APIs, coarse keyword filter
│ └── job_candidates.json # raw candidates (gitignored)
│
└── Claude (SKILL.md) # JUDGE: reads candidates, scores fit,
# presents shortlist, drives triage
/resume-cover <url>
│
├── WebFetch # Fetch & parse the job posting
├── Source resume (.docx) # Read master resume as truth
├── Claude # Tailor content to the JD
└── generate_resume_cover.py # Build .docx output files
License
MIT