# test-prompt

Runs your prompt scenario with and without the prompt on a subagent, so you can see exactly what the instruction changes.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 1200 installs · ★ 1787
- Author: NeoLabHQ
- License: GPL-3.0 · policy=link-only · description only; the file stays with the author
- Source: https://github.com/NeoLabHQ/context-engineering-kit/tree/HEAD/skills/test-prompt/SKILL.md
- Updated: 2026-08-26
- HTML: https://affpapa.org/en/nevibecode/skills/test-prompt

## What it does

The skill borrows the RED-GREEN-REFACTOR cycle from test-driven development. First you run a scenario with no prompt loaded and write down how the agent behaves on its own. Then you draft an instruction aimed at those failures and rerun the same scenario with it. The last pass tightens weak spots and checks the result on a fresh subagent.

It covers slash commands, hooks, skills, subagent instructions and prompts for production models. A subagent starts clean, with no chat history, so only the prompt under test affects the outcome. For rules meant to hold under pressure, the scenario adds it: a looming deadline, fatigue, working code you would hate to throw away.

The skill expects `test-driven-development` and `prompt-engineering` as background. The process uses the Task tool with `general-purpose`, and the subagent must not see the prompt being tested — otherwise the baseline run proves nothing.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add NeoLabHQ/context-engineering-kit --skill test-prompt -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/test-prompt/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/test-prompt/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add NeoLabHQ/context-engineering-kit --skill test-prompt -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/test-prompt/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add NeoLabHQ/context-engineering-kit --skill test-prompt -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/test-prompt/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install NeoLabHQ/context-engineering-kit test-prompt
```

2. **Manually: put the files in place**
- File: `.github/skills/test-prompt/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add NeoLabHQ/context-engineering-kit --skill test-prompt -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/NeoLabHQ/context-engineering-kit --path skills/test-prompt
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/test-prompt/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add NeoLabHQ/context-engineering-kit --skill test-prompt -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/NeoLabHQ/context-engineering-kit/tree/HEAD/skills/test-prompt/SKILL.md

## When to use it
- record how an agent fails without the instruction
- draft a prompt that targets the failures you observed
- test a slash command on a fresh subagent before release
- find loopholes an agent uses to dodge a rule
- compare two prompt versions on the same scenario

## Example prompts
- Run the commit scenario without my command and log what the agent gets wrong
- Test this skill on a subagent, first without it and then with it
- Tighten the hook prompt and recheck it on a new subagent

## Similar
- [setup-matt-pocock-skills](https://affpapa.org/en/nevibecode/skills/setup-matt-pocock-skills.md) — skill · 1052851 installs
- [ask-matt](https://affpapa.org/en/nevibecode/skills/ask-matt.md) — skill · 766587 installs
- [to-tickets](https://affpapa.org/en/nevibecode/skills/to-tickets.md) — skill · 711230 installs
- [writing-plans](https://affpapa.org/en/nevibecode/skills/writing-plans.md) — skill · 282859 installs
- [caveman-review](https://affpapa.org/en/nevibecode/skills/caveman-review.md) — skill · 411800 installs
- [caveman-compress](https://affpapa.org/en/nevibecode/skills/caveman-compress.md) — skill · 411500 installs

From the same set: [neolabhq/context-engineering-kit](https://affpapa.org/en/nevibecode/by/neolabhq/context-engineering-kit.md) — 64 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md