widescreen-research: Command for Claude Code

.claude/commands/games/sandbox-testing-mechanism.md

sandbox-testing-mechanism is a command for Claude Code from glassBead-tc/widescreen-research. It costs 0 tokens per session (2,864 once invoked), scanned A, original, MIT.

A command for testing newly generated code in a temporary isolated environment called a sandbox. It builds a plausible example or application, runs tests, captures results, and cleans up afterward.

In plain words
What is it for?
Use it to verify documentation, software-development kits, or open-source contributions by building and testing them safely.
Why use it?
It can reveal compilation, runtime, or usage problems before the code is relied on, without changing the code under test.

Command for Claude Code

Written for Claude Code: installed under .claude/.

This is glassBead-tc/widescreen-research's own configuration. It tells Claude Code how to work on widescreen-research itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything widescreen-research configures →

Reuse

Borrowing it

Nothing to install: this file belongs to glassBead-tc/widescreen-research. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/glassBead-tc/widescreen-research/main/.claude/commands/games/sandbox-testing-mechanism.md
Clone the repo
git clone --depth 1 https://github.com/glassBead-tc/widescreen-research

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for sandbox-testing-mechanism

README.md
[![agentmods](https://agentmods.dev/badge/commands/glassbead-tc/widescreen-research/sandbox-testing-mechanism/github.svg)](https://agentmods.dev/commands/glassbead-tc/widescreen-research/sandbox-testing-mechanism)
Your own site
<a href="https://agentmods.dev/commands/glassbead-tc/widescreen-research/sandbox-testing-mechanism"><img src="https://agentmods.dev/badge/commands/glassbead-tc/widescreen-research/sandbox-testing-mechanism/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for sandbox-testing-mechanism

Your own site · 80×15
<a href="https://agentmods.dev/commands/glassbead-tc/widescreen-research/sandbox-testing-mechanism"><img src="https://agentmods.dev/badge/commands/glassbead-tc/widescreen-research/sandbox-testing-mechanism.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,864 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00000 $0.02864
Opus 5 $0.00000 $0.01432
Sonnet 5 $0.00000 $0.00573
Haiku 4.5 $0.00000 $0.00286

Measured 10d ago against content hash 6b2511f00ce4, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

sandbox-testing-mechanism scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/commands/games/sandbox-testing-mechanism.md · 233 lines

How it starts

The opening of the file, as written. The whole thing — 233 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Sandbox Testing Mechanism

Test newly generated code inside an E2B sandbox (or container-use) by building a plausible application or demonstration. This Stage-Gate workflow forces the agent to compile, test and use its own code in a safe, disposable environment and exposes failures early.

Overview

  • Goal - Verify that freshly written documentation, SDK code or OSS contribution actually works by building something with it in a sandbox.
  • Pattern - Stage-Gate phases with explicit go/kill gates and anti-pattern detection. Use E2B Sandboxes and the Model Context Protocol (MCP) server to create a virtual desktop, run commands, and capture results. E2B sandboxes let AI agents execute code in a secure isolated environment and provide actions such as create_sandbox, execute_computer_action, get_screenshot and cleanup_sandbox.
  • Scope - Any codebase (TypeScript, Python, etc.). The workflow is additive: it does not alter the code under test and cleans up resources when done.

Usage

/sandbox-testing-game $CODE_SOURCE [build_command] [test_command] [example_goal] [timeout]

Arguments

  • $CODE_SOURCE (required): Path or URL to the code to test. Accepts a Git repository URL, local directory/file, or inline snippet. For Git URLs the code is cloned; for local paths it is copied; for inline snippets it is written to a file.
  • build_command (optional): Shell command to build or install the code. If omitted the workflow infers a default (e.g. npm install for Node projects or pip install . for Python).
  • test_command (optional): Shell command to run automated tests. Defaults are inferred (e.g. npm test, pytest -q).
  • example_goal (optional): Plain-text description of a simple but real example that exercises the code (e.g. “build a CLI that lists items” or “display a chart”). If omitted the workflow uses the project’s quickstart or README.
  • timeout (optional): Maximum sandbox lifetime in milliseconds. Default: 600000 (10 minutes).

Read the full file on GitHub · 233 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 10d ago First seen · 233 lines · 0 tokens per session scan A 6b2511f00ce4

Subscribe to this mod's changes

sandbox-testing-mechanism is a command published in the GitHub repository glassBead-tc/widescreen-research (6 stars, last pushed 10mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,864 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.