Prompts tested against real agent behavior.
What it does
Guides agents through isolated tests and revisions to produce prompts that correct observed failures.
When to use it
Use when creating or editing prompts, including commands, hooks, skills, and subagent instructions.
How to use it
It changes how the agent creates and edits prompts by requiring isolated baseline tests, targeted revisions, and robustness checks.
What you provide
- An existing project
Good to know
- starts helper agents, number unstated
- starts 2 helper agents
Latest review
What you need · 1
Requires understanding this skill's RED-GREEN-REFACTOR cycle as background; installation is not explicitly required.
Access · 0
No additional actions listed.