Claude Code skills and browser userscripts for LinkedIn data analysis, career visualization, and live metrics capture. Turn a LinkedIn data export into interactive dashboards, network graphs, and career timelines — or capture your post analytics as you browse. All from your terminal and your browser, no servers.
git clone https://github.com/thatrebeccarae/linkedin-toolkit.gitLive Demo · Why I Built This · Who This Is For · Getting Started · Skills · Userscripts · License
LinkedIn gives you a data export, but no tools to make sense of it. You get CSVs full of connections, messages, invitations, and inferences — but no way to see patterns, identify blind spots, or understand how your network actually works.
I wanted to answer questions like: Who should I be talking to more? Is posting actually driving connection growth? Which parts of my network are strongest? LinkedIn's own analytics don't go deep enough, and most third-party tools want your login credentials.
LinkedIn Toolkit keeps everything local. Your data never leaves your machine. Claude Code runs the analysis, Python generates the visualizations, and you open the results in your browser. No servers, no accounts, no API keys required.
- Professionals who want to understand their LinkedIn network beyond surface-level stats
- Job seekers who want to identify warm connections at target companies
- Content creators who want to know if posting actually drives growth
- Posters who want their LinkedIn post metrics auto-exported to a spreadsheet or webhook
- Career changers who want to visualize their professional evolution
- Anyone who requested a LinkedIn data export and has no idea what to do with it
git clone https://github.com/thatrebeccarae/linkedin-toolkit.git
cp -r linkedin-toolkit/skills/linkedin-data-viz/ ~/.claude/skills/Then open Claude Code and say: "Analyze my LinkedIn data export"
The skill walks you through a wizard: locating your export, previewing your data, selecting a theme, and generating visualizations. No configuration needed.
Don't have a LinkedIn data export yet?
- Go to LinkedIn Settings > Data Privacy
- Click Get a copy of your data
- Select all categories (or at minimum: Connections, Messages, Invitations, Company Follows)
- Click Request archive — LinkedIn emails you a download link within ~24 hours
- Download and unzip the archive
- Python 3.9+ (no pip installs needed — stdlib only)
- Claude Code with skills support
- Tampermonkey (Chrome/Firefox/Safari) — only if you want the userscript
LinkedIn Data Viz — 10 Interactive Visualizations
View Live Demo — See all 10 visualizations with sanitized sample data.
Turn a LinkedIn data export into a complete visual analysis:
| Visualization | What It Shows |
|---|---|
| Network Universe | D3.js force-directed graph of connections clustered by role |
| Inferences vs Reality | LinkedIn's algorithmic guesses about you vs actual data |
| High-Value Messages | Unanswered messages ranked by potential value |
| Company Follows | Followed companies grouped by industry |
| Inbound vs Outbound | Connection request patterns over time |
| Connection Quality | Relationship depth analysis (Active / Dormant / Never Messaged) |
| Connection Timeline | Recent connections with engagement data |
| Inbox Quality | Genuine conversations vs noise over time |
| Does Posting Work? | Posting frequency correlated with connection growth |
| Career Strata | Your network as geological layers by career phase |
Plus a unified dashboard linking all visualizations together.
All data stays local. No API keys. No servers. Just Python, your browser, and your LinkedIn export.
Sanitize for sharing
Want to share your visualizations publicly? The built-in sanitizer replaces real names and companies with plausible fakes — all patterns stay intact but your contacts' privacy is protected.
LinkedIn Metrics Collector — Passive post analytics capture
A Tampermonkey userscript that quietly captures your LinkedIn post analytics as you browse your own analytics pages — impressions, reactions, comments, shares, the whole breakdown — and saves them locally or posts them to any webhook you control (n8n, Zapier, a Cloudflare Worker, your own server). No login, no scraping, nothing your browser wasn't already fetching.
It doesn't log in for you, doesn't touch other users' data, doesn't use a headless browser. It only reads API responses LinkedIn is already serving to your own browser, and stores them somewhere you can actually use them. GDPR Art. 20 data portability, own data only.
Bug reports, documentation fixes, and new skill ideas are welcome. Please open an issue first for new skills so we can discuss scope and approach.
MIT License. See LICENSE for details.
Your LinkedIn data, visualized. Install the skill. Say "analyze my LinkedIn data export." See your network clearly for the first time.