Do I need an E2B API key to use this skill?

No, an E2B API key is only required if you want to use the Cloud runner. Local testing is the default recommended mode and requires no additional setup.

What metrics does the Sandbox Test Runner provide?

It provides a detailed summary including pass/fail status, execution duration, token usage, estimated cost, and specific error logs for failed steps.

Can I test individual skills or commands?

Yes, the interactive configuration allows you to target specific skills or commands, making it ideal for debugging individual components of your workflow.

How long do the tests usually take?

P0 smoke tests typically take around 10 minutes for 10 tests, while a full suite of 25 tests may take up to an hour depending on the complexity of the skills.

What is a sandbox test in Claude Code?

A sandbox test executes a specific skill or command in a controlled, isolated environment to verify its behavior, tool calls, and outputs without affecting your main workspace.

Sandbox Test Runner

Name: Sandbox Test Runner
Author: jrc1883

byjrc1883

Security & Testing

Executes interactive sandbox tests for Claude Code skills and commands in isolated environments to validate functionality and prevent regressions.

About

The Sandbox Test Runner provides a comprehensive testing platform for the PopKit ecosystem, allowing developers to run isolated smoke tests or full suites on skills and commands. It supports both local and cloud-based execution via E2B, capturing detailed telemetry, tool traces, and performance metrics to ensure high-quality AI workflows. By offering comparative analysis and automated reporting, it streamlines the debugging and validation process for complex AI-driven development tools, providing insights into token usage, decision points, and execution timing.

Key Features

Isolated execution environments via Local or E2B Cloud runners
0 GitHub stars
Regression detection through comparative result analysis
Interactive test selection for smoke tests, full suites, or specific skills
Automated performance reporting with token usage and cost analysis
Comprehensive telemetry capturing tool traces and decision points

Use Cases

Validating new custom skills in isolation before deployment
Analyzing AI decision-making patterns through detailed tool traces
Running P0 smoke tests to ensure core plugin functionality

About

Key Features

Isolated execution environments via Local or E2B Cloud runners
0 GitHub stars
Regression detection through comparative result analysis
Interactive test selection for smoke tests, full suites, or specific skills
Automated performance reporting with token usage and cost analysis
Comprehensive telemetry capturing tool traces and decision points

Use Cases

Validating new custom skills in isolation before deployment
Analyzing AI decision-making patterns through detailed tool traces
Running P0 smoke tests to ensure core plugin functionality