smithery.ai

website-preview

Take screenshots and interact with the rendered Jekyll website

First seen Mar 22, 2026

Installation

$ npx skills add https://smithery.ai

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

Also in this package

Other skills from smithery.ai · top by installs.

npx skills add https://smithery.ai

Browse all from smithery.ai

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Not declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 3,372 B
  • docs SUMMARY.md 85 B

History

  1. First seen on skills.sh
  2. First recorded snapshot · 2 installs

SKILL.md

Website Preview

Render the Jekyll site locally, take screenshots, and verify design changes visually using ChromeDriver and browser automation.

When to use this skill

Use this skill when:

  • Modifying CSS, SCSS, or any styling files
  • Editing HTML layouts (layouts/) or includes (includes/)
  • Making changes that affect visual appearance
  • The user asks to see how changes look
  • You need to verify a design change worked correctly
  • Working on any front-end aspects of the website

Invoke this skill proactively after making visual changes to the site.

How to use this skill

Step 1: Check Prerequisites

Before taking screenshots, verify both Jekyll and ChromeDriver are running.

Check Jekyll server:

curl -s -o /dev/null -w "%{http_code}" http://localhost:4000/ 2>/dev/null || echo "not running"

If not running (not 200), start it in background:

bundle exec jekyll serve --host 0.0.0.0 &

Wait a few seconds for it to build.

Check ChromeDriver:

cat /tmp/chromedriver_port 2>/dev/null || echo "not running"

If not running, start it:

./browser.sh chromedriver

This finds Chrome automatically and starts ChromeDriver on a free port.

Step 2: Take Screenshots

Use ./browser.sh for browser automation:

./browser.sh start http://localhost:4000/
./browser.sh screenshot /tmp/preview.png

./browser.sh dark
./browser.sh screenshot /tmp/preview-dark.png

./browser.sh mobile
./browser.sh screenshot /tmp/preview-mobile-dark.png

./browser.sh light
./browser.sh screenshot /tmp/preview-mobile-light.png

./browser.sh desktop
./browser.sh nav http://localhost:4000/2026/course-shell/
./browser.sh screenshot /tmp/course-shell.png

./browser.sh stop

Mode commands (dark, light, mobile, desktop) persist until changed.

Step 3: View Screenshots

After taking a screenshot, use the Read tool to view it:

Read /tmp/preview.png

Inspecting Elements

Execute JavaScript to query the page:

./browser.sh js "document.querySelector('nav').getBoundingClientRect()"

After Making Changes

After modifying CSS/HTML:

  1. Jekyll will auto-rebuild (watch for "Regenerating:" in Jekyll output)
  2. Refresh: ./browser.sh nav http://localhost:4000/
  3. Take a new screenshot to verify changes

Common Pages to Check

  • Homepage: http://localhost:4000/
  • Current year index: http://localhost:4000/2026/
  • Lecture page: http://localhost:4000/2026/course-shell/
  • About page: http://localhost:4000/about/

ChromeDriver Lifecycle

Keep ChromeDriver running across multiple previews during a session. Only kill it when no more browser interactions are expected.

  • ./browser.sh stop - Ends the browser session (optional between screenshots)
  • ./browser.sh kill - Stops ChromeDriver entirely (only when done with all design work)

You do not need to run ./browser.sh kill after each preview. Start ChromeDriver once at the beginning and leave it running.

Troubleshooting

Jekyll not rebuilding: Run bundle exec jekyll build manually

ChromeDriver errors: Kill and restart

./browser.sh kill
./browser.sh chromedriver

Session errors: Stop any existing session first

./browser.sh stop