cmd-skills-dashboard
Skills Dashboard Generator
Scrape the skills.sh registry and produce an interactive Plotly.js dashboard showing who publishes the most skills, who has the most installs, and how adoption is distributed.
When to Use This Skill
- Generating a fresh snapshot of the skills.sh ecosystem
- Comparing publishers by skill count vs install count
- Exploring the power-law distribution of skill adoption
- Answering "who dominates the skills ecosystem?"
How It Works
- Scrape - Fetch
https://skills.sh/api/searchwith broad 2-char queries to discover all skills - Aggregate - Group by owner (GitHub org/user) and repo, compute counts and totals
- Render - Generate a self-contained HTML file with Plotly.js charts
Running
Generate the dashboard with the scraper script:
python3 scripts/scrape_and_build.py
This writes index.html to the current directory.
To write to a specific path:
python3 scripts/scrape_and_build.py --output /path/to/dashboard.html
Dashboard Contents
| Chart | What It Shows |
|---|---|
| Treemap | Install share by publisher, click to drill into individual skills |
| Bar: Skill Count | Top 25 publishers by number of skills published |
| Bar: Total Installs | Top 25 publishers by total install volume |
| Histogram | Log-scale distribution showing the long tail of installs |
| Top 30 Skills | The 30 most-installed individual skills |
Data Source
All data comes from the skills.sh/api/search endpoint. Each entry has:
{
"source": "owner/repo",
"skillId": "skill-name",
"name": "skill-name",
"installs": 12345
}
Generate It Yourself
Install the skill:
npx skills add olshansk/agent-skills
Select skills-dashboard when prompted.
Then:
- Launch your agent CLI of choice (Claude Code, Codex, Gemini CLI, OpenCode)
- Ask it to "build the skills ecosystem dashboard"
No API keys needed — the skill scrapes the public skills.sh registry and outputs a self-contained HTML file.
More from olshansk/agent-skills
session-commit
Capture learnings from the current coding session and update AGENTS.md. Use when the user asks to close the loop, run session-commit, record best practices, or update agent instructions based on recent work.
30skills-dashboard
Scrape skills.sh and generate an interactive HTML dashboard showing skill distribution by publisher, installs, and categories. Rerun anytime to get fresh data.
29cmd-clean-code
Improve code readability without altering functionality using idiomatic best practices
25cmd-idiot-proof-docs
Simplify documentation for clarity and scannability with approval-gated edits
18cmd-rss-feed-generator
Generate Python RSS feed scrapers from blog websites, integrated with hourly GitHub Actions
18cmd-proofread
Proofread posts before publishing for spelling, grammar, repetition, logic, weak arguments, broken links, and optionally reformat for skimmability or shape the writing vibe toward a known author's style
17