Skip to content
Get packs

Vercel Sandbox Browser Automation

Lets an agent run headless Chrome browser tasks inside disposable Vercel cloud sandboxes from Vercel-hosted apps.

Source Download skill
Repository
vercel-labs/agent-browser
Stars
45,202
Licence
Apache-2.0
Files
1 file

What it does

This skill teaches an AI agent how to control a real headless Chrome browser inside a temporary cloud virtual machine on Vercel. The agent can open web pages, take screenshots, read an accessibility snapshot of a page, and run multi-step actions like filling and submitting forms, then the VM shuts down. It is written for developers building Vercel-deployed apps (Next.js, SvelteKit, Nuxt, Remix, Astro and similar) who need browser automation on the server side, including scheduled jobs via Vercel Cron.

Great for

  • Developers adding server-side browser automation, such as page screenshots or content checks, to a Vercel-hosted web app
  • Teams who want scheduled browser tasks (e.g. monitoring a pricing page daily) run through Vercel Cron
  • Engineers who need isolated, ephemeral browser environments that do not persist between runs

Skip it if

  • Non-developers or people without a Vercel account, since it depends on Vercel Sandbox and its credentials
  • Automating sites you do not have permission to access, since the skill performs real logins and form submissions that may breach terms of service
  • Tasks needing a browser on your own machine; this skill is specific to Vercel infrastructure

Try asking

  • “Take a screenshot of https://example.com inside my Next.js app using Vercel Sandbox”
  • “Get an accessibility snapshot of our pricing page from a Vercel cron job every morning”
  • “Fill and submit the signup form on this URL in a sandboxed Chrome and capture the confirmation screen”
Step-by-step guide: how to use Vercel Sandbox Browser AutomationInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the Apache-2.0 licence and an attribution note.

    vercel-sandbox.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip vercel-sandbox.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip vercel-sandbox.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Create a sandbox snapshot once and set AGENT_BROWSER_SNAPSHOT_ID, otherwise each run spends about 30 seconds installing dependencies
  • Keep Vercel tokens in environment variables and never hard-code them; use a narrowly scoped token for local development
  • Review any form-submission or login automation before running it against live sites, and only target sites you are authorised to automate

What's inside

  • SKILL.md

Redistributed unmodified from vercel-labs/agent-browser under the Apache-2.0 licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.