tsmc-resume-matcher

A Claude Code skill that matches your resume against job postings using AI. Focused on TSMC careers, but works with any job URL.

What it does

Paste your resume (PDF or text) and one or more job URLs. The skill:

1. Parses your resume into structured fields
2. Scrapes the job description (uses Playwright for TSMC's Cloudflare-protected SPA; standard fetch for other sites)
3. Scores the match across 5 dimensions (100 pts total)
4. Outputs a report with score, grade, strengths, gaps, and recommendations

Scoring dimensions

Dimension	Points
Skills match (required + preferred)	35
Experience level	20
Industry background	20
Education	10
Soft skills / culture fit	15

Grades: S (90+) / A (75–89) / B (60–74) / C (45–59) / D (<45)

Requirements

• Claude Code
• Python 3.11+ (for the TSMC scraper)

Installation

git clone https://github.com/PttCodingMan/tsmc-resume-matcher.git
cd tsmc-resume-matcher
./install.sh

The skill is installed to ~/.claude/skills/tsmc-resume-matcher/.

Set up the TSMC scraper (optional)

Only needed if you want to scrape TSMC job listings. Skippable if you paste job descriptions directly.

cd ~/.claude/skills/tsmc-resume-matcher
python3 -m venv .venv
.venv/bin/pip install playwright
.venv/bin/playwright install chromium

Screenshots

Skill launch

JD parsing

Final job ranking

Usage

In any Claude Code session, invoke the skill:

/tsmc-resume-matcher

Then follow the prompts — attach your resume PDF or paste the text, and provide job URLs or let the scraper find TSMC listings interactively.

Project structure

tsmc-resume-matcher/
├── SKILL.md                  # Skill instructions loaded by Claude Code
├── scraper.py                # Playwright scraper for TSMC careers
└── references/
    ├── resume-parser.md      # Resume parsing rules
    ├── jd-parser.md          # Job description parsing rules
    ├── scoring-rubric.md     # Scoring rubric details
    └── tsmc-scraper.md       # Scraper usage notes

License

MIT