What is the RED phase in this context?

The RED phase involves identifying a real execution failure—such as a commit to the wrong branch—and creating a test scenario that reproduces that specific failure.

What is workflow testing with subagents?

It is the process of using AI subagents to test the clarity of your instructions by trying to break them under simulated pressure to find ambiguities.

Why use simulated pressure during testing?

Agents are more likely to take shortcuts or misread steps when simulated as 'tired' or 'rushed,' revealing hidden ambiguities that wouldn't appear in perfect conditions.

When should I use this skill?

Use it whenever you are creating complex multi-step commands or orchestrator prompts where the specific order of execution is critical for the task's success.

How does this differ from testing-skills?

While testing-skills focuses on whether an agent will comply with rules, workflow testing focuses on whether the agent can correctly understand and follow specific task steps.

Workflow Testing with Subagents

Name: Workflow Testing with Subagents
Author: arittr

byarittr

•

安全与测试

Validates the clarity of orchestrator prompts and workflow documentation by simulating real-world execution failures with subagents.

关于

This skill adapts Test-Driven Development (TDD) principles to the creation and optimization of Claude Code workflow instructions and command documentation. By identifying real-world execution failures, creating reproducible test scenarios under simulated pressure, and using subagents to verify fixes, it ensures that complex multi-step instructions are unambiguous and resistant to misinterpretation. It is particularly useful for developers building complex orchestrator prompts or multi-branch git workflows where clarity and execution order are critical for success, providing a structured framework to move from failing execution (RED) to verified clarity (GREEN).

主要功能

Applies the Red-Green-Refactor cycle to non-code workflow documentation
Identifies root causes of subagent misinterpretation in logs and reports
Facilitates meta-testing to diagnose why previous instruction fixes failed
Simulates realistic execution pressure to find instruction loopholes
6 GitHub stars
Provides specific fix patterns for explicit sequential ordering and warnings

使用场景

Troubleshooting recurring errors in automated git workflows
Hardening command documentation against agent misinterpretation
Refining orchestrator prompts for complex multi-agent tasks

关于

主要功能

Applies the Red-Green-Refactor cycle to non-code workflow documentation
Identifies root causes of subagent misinterpretation in logs and reports
Facilitates meta-testing to diagnose why previous instruction fixes failed
Simulates realistic execution pressure to find instruction loopholes
6 GitHub stars
Provides specific fix patterns for explicit sequential ordering and warnings

使用场景

Troubleshooting recurring errors in automated git workflows
Hardening command documentation against agent misinterpretation
Refining orchestrator prompts for complex multi-agent tasks