Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
alsk1992 avatar

Sandbox

  • 12 installs
  • 610 repo stars
  • Updated June 26, 2026
  • alsk1992/cloddsbot

Sandbox is a Claude Code skill for the clodds bot that pushes HTML to a live-updating canvas server for visual previews and agent-driven UI.

About

Sandbox is a clodds skill that serves a local HTTP canvas page and renders pushed HTML in real time. Developers use it to give an agent a visual surface for previews, dashboards, and agent-driven UI, and to capture screenshots of the current state. It matters when a chat agent needs to show rendered output rather than plain text.

  • Pushes HTML to a live-updating canvas server for visual previews and dashboards
  • Chat commands to start, push, reset, screenshot, and check status of the canvas
  • JavaScript eval and unsafe mode are off by default via env flags

Sandbox by the numbers

  • 12 all-time installs (skills.sh)
  • Ranked #1,643 of 2,245 Frontend Development skills by installs in the Skillselion catalog
  • Data as of Aug 5, 2026 (Skillselion catalog sync)
At a glance

sandbox capabilities & compatibility

Capabilities
html preview · screenshot · dashboard
Use cases
ui design
Pricing
Free
From the docs

What sandbox says it does

Push HTML content to a live-updating canvas server for visual previews, dashboards, and agent-driven UI.
SKILL.md
`/sandbox screenshot` captures the current canvas state as an image
SKILL.md
npx skills add https://github.com/alsk1992/cloddsbot --skill sandbox

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
Installs12
repo stars610
Last updatedJune 26, 2026
Repositoryalsk1992/cloddsbot

What it does

Push HTML to a live canvas so an agent can render a visual preview or dashboard and screenshot it.

Who is it for?

Giving a chat agent a live visual canvas to render HTML previews, dashboards, or UI.

Skip if: Production web hosting or persistent front-end applications.

When should I use this skill?

You want an agent to render and screenshot HTML output instead of returning text.

What you get

HTML is rendered on a live canvas and can be screenshotted.

  • live canvas url
  • rendered html preview
  • canvas screenshot

By the numbers

  • 5 canvas commands (start, push, reset, screenshot, status)
  • 2 env flags (CANVAS_ALLOW_JS_EVAL, ALLOW_UNSAFE_SANDBOX)

Files

SKILL.mdMarkdownGitHub ↗

Sandbox - Visual Canvas

Push HTML content to a live-updating canvas server for visual previews, dashboards, and agent-driven UI.

Commands

/sandbox start        - Start the canvas server (returns URL)
/sandbox push <html>  - Push HTML content to the canvas
/sandbox reset        - Reset canvas to blank state
/sandbox screenshot   - Take a screenshot of the current canvas
/sandbox status       - Show canvas server status and settings

Environment Variables

VariableDescription
CANVAS_ALLOW_JS_EVALEnable JavaScript evaluation in canvas (default: false)
ALLOW_UNSAFE_SANDBOXEnable unsafe sandbox mode (default: false)

Examples

/sandbox start
/sandbox push <h1>Hello World</h1><p>Live preview</p>
/sandbox screenshot
/sandbox status
/sandbox reset

How It Works

1. /sandbox start launches a local HTTP server that serves the canvas page 2. /sandbox push <html> sends HTML to the canvas, rendered in real time 3. /sandbox screenshot captures the current canvas state as an image 4. /sandbox reset clears all HTML, CSS, and JS from the canvas 5. /sandbox status shows whether the server is running, the URL, and content sizes

Related skills

FAQ

Is JavaScript executed in the canvas?

No by default. It is enabled only when CANVAS_ALLOW_JS_EVAL is set.

How do I capture the current canvas?

Run /sandbox screenshot to capture the current canvas state as an image.

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.