# karpathy-jobs-bls-visualizer

This skill scrapes Occupational Outlook Handbook data, scores occupations with an LLM and renders an interactive jobs treemap.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 1300 installs · ★ 88
- Author: reason-machines
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/reason-machines/trending-skills/tree/HEAD/skills/karpathy-jobs-bls-visualizer/SKILL.md
- Updated: 2026-10-09
- HTML: https://affpapa.org/en/nevibecode/skills/karpathy-jobs-bls-visualizer

## What it does

The skill covers the whole path from raw Bureau of Labor Statistics pages to a colored map: scraping, converting HTML to Markdown, extracting fields into `occupations.csv`, scoring with an LLM, and merging everything into `site/data.json` for the frontend. Rectangle area shows employment size, color shows whichever metric you pick: growth outlook, median pay, education, or an AI exposure score.

The scoring layer is the part worth forking. Edit the prompt in `score.py`, run `uv run python score.py`, then `uv run python build_site_data.py` and the treemap gets a new color. Adding a layer means a toggle button in `site/index.html` plus a branch in the color function.

LLM scoring needs an OpenRouter key in a `.env` file. Scraping runs Playwright with a visible browser window because BLS blocks bots, so `scrape.py` will not work headless or on a server without a graphical environment. Dependencies come from `uv sync`, plus `uv run playwright install chromium`.

## Requirements
- Works with: Playwright

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill karpathy-jobs-bls-visualizer -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/karpathy-jobs-bls-visualizer/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/karpathy-jobs-bls-visualizer/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill karpathy-jobs-bls-visualizer -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/karpathy-jobs-bls-visualizer/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill karpathy-jobs-bls-visualizer -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/karpathy-jobs-bls-visualizer/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install reason-machines/trending-skills karpathy-jobs-bls-visualizer
```

2. **Manually: put the files in place**
- File: `.github/skills/karpathy-jobs-bls-visualizer/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill karpathy-jobs-bls-visualizer -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/reason-machines/trending-skills --path skills/karpathy-jobs-bls-visualizer
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/karpathy-jobs-bls-visualizer/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill karpathy-jobs-bls-visualizer -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/reason-machines/trending-skills/tree/HEAD/skills/karpathy-jobs-bls-visualizer/SKILL.md

## When to use it
- map 342 occupations from BLS data
- re-score AI exposure with your own prompt
- add a new color layer to the jobs treemap
- pack all occupation data into one file for an LLM chat
- turn raw BLS pages into clean Markdown

## Example prompts
- run the full karpathy-jobs-bls-visualizer pipeline and build site/data.json
- rewrite the prompt in score.py for robotics exposure and re-run scores.json
- add a robotics_score layer to the treemap and update build_site_data.py

## Similar
- [python-executor](https://affpapa.org/en/nevibecode/skills/python-executor.md) — skill · 109700 installs
- [web-scraping](https://affpapa.org/en/nevibecode/skills/web-scraping.md) — skill · 6100 installs
- [extract](https://affpapa.org/en/nevibecode/skills/actionbook-extract.md) — skill · 2300 installs
- [gemini-computer-use](https://affpapa.org/en/nevibecode/skills/gemini-computer-use.md) — skill · 1200 installs
- [playwright-cli](https://affpapa.org/en/nevibecode/skills/playwright-cli.md) — skill · 195700 installs
- [anti-detect-browser](https://affpapa.org/en/nevibecode/skills/anti-detect-browser.md) — skill · 137300 installs

From the same set: [reason-machines/trending-skills](https://affpapa.org/en/nevibecode/by/reason-machines/trending-skills.md) — 42 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md