benchmark-sandbox

Benchmark Vercel plugins in Sandboxes and generate structured reports.

Updated May 31, 2026
One-click install
npx skills add https://github.com/Extremez-Surya/onlineruler --skill benchmark-sandbox-extremez-surya
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: benchmark-sandbox
Source: https://github.com/Extremez-Surya/onlineruler/tree/main/.agents/skills/benchmark-sandbox
Command: npx skills add https://github.com/Extremez-Surya/onlineruler --skill benchmark-sandbox-extremez-surya

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @vercel/sandbox, claude-code, agent-browser, vercel-cli, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates the performance benchmarking of Vercel plugins by running them inside ephemeral Vercel Sandboxes, saving time and resources compared to traditional methods.

Core Features & Use Cases

  • Remote Plugin Testing: Runs plugin code in Vercel Sandboxes instead of local environments.
  • Ephemeral MicroVMs: Utilizes Firecracker microVMs for isolated testing.
  • Automated Evaluation: Includes 3-phase pipeline: Build, Verify, Deploy, with automated scoring.
  • Use Case: Use this Skill to quickly evaluate the performance and functionality of a new Vercel plugin by running a predefined benchmark and obtaining a structured report.

Quick Start

Run the benchmark for the plugin 'my-plugin' using the following command:

bun run .claude/skills/benchmark-sandbox/run-eval.ts --scenarios my-plugin

Frequently Asked Questions about benchmark-sandbox

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate performance benchmarking for Vercel plugins?

Automate performance benchmarking for Vercel plugins by provisioning ephemeral Vercel Sandboxes to run the plugin code, utilizing Claude Code for automated evaluation across Build, Verify, and Deploy phases to generate structured reports.

What is the best way to test Vercel plugins in an isolated environment?

Testing Vercel plugins in an isolated environment is achieved by utilizing Firecracker microVMs within Vercel Sandboxes, which provide ephemeral infrastructure to securely execute and evaluate plugin code without affecting local setups.

Do I need Vercel CLI and Claude Code to run sandbox benchmarks?

Yes, running sandbox benchmarks requires Vercel CLI, Claude Code, agent-browser, and the Vercel Sandbox dependency to provision the microVMs, execute the plugin scenarios, and automate the evaluation pipeline.

How does the automated evaluation pipeline work for Vercel plugin testing?

The automated evaluation pipeline works by executing a 3-phase process—Build, Verify, and Deploy—within Vercel Sandboxes, applying automated scoring to assess plugin performance and functionality before generating a final structured report.

Can I run a specific Vercel plugin benchmark scenario from the command line?

Yes, you can run a specific Vercel plugin benchmark scenario by executing the run-eval.ts script via Bun, passing the target plugin name as the scenarios argument to trigger the automated testing pipeline.

Why use Vercel Sandboxes instead of local environments for plugin performance testing?

Vercel Sandboxes are used instead of local environments for plugin performance testing to save time and resources, leveraging ephemeral Firecracker microVMs to ensure consistent, isolated execution without local configuration overhead.