# message-test-designer

Designs the message test before you spend on scale: hypothesis, panel, protocols, success thresholds and a stop-or-revise rule.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 1000 installs · ★ 3013
- Author: aaron-he-zhu
- License: Apache-2.0 · policy=mirror · the beginning of the file is at the end of this page
- Source: https://github.com/aaron-he-zhu/aaron-marketing-skills/tree/HEAD/narrative/evaluate/message-test-designer/SKILL.md
- Updated: 2026-10-25
- HTML: https://affpapa.org/en/nevibecode/skills/message-test-designer

## What it does

The skill takes a candidate message — a tagline, one-liner, value pillars or a per-surface message-match spec — and turns it into a test design spec. Inside: the hypothesis, panel and recruit criteria, comprehension and 5-second recall protocols, a Wynter-style message-market-fit panel, a stimulus set drawn from the canon, success thresholds, and a stop/revise decision rule. It is the design layer of the TALE Evaluate phase.

Design only. The skill does not run the panel or the A/B experiment — execution goes to the experiment builders, result analysis to performance-analyzer, and edits to the message itself to message-system-architect. Unverifiable claims in a stimulus are marked `[needs source]` and submitted to `memory/events/claims.ndjson` via an `operation: propose` request to `registry-events.py`; offer-claims-registry is the sole adjudicator.

A result cannot be applied unless its shared `evidence-observation` references the same Narrative Truth, Stimulus and Retro binding. Change the canon, stimulus, panel, protocol or contract and a new test starts. The spec lands in `memory/narrative/message-test-designer/`, while the chosen hypothesis and pass thresholds go to `memory/open-loops.md` as a pending-decision item — the skill asks before writing.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add aaron-he-zhu/aaron-marketing-skills --skill message-test-designer -a claude-code
```

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

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add aaron-he-zhu/aaron-marketing-skills --skill message-test-designer -a codex
```

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

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add aaron-he-zhu/aaron-marketing-skills --skill message-test-designer -a cursor
```

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

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install aaron-he-zhu/aaron-marketing-skills message-test-designer
```

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

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add aaron-he-zhu/aaron-marketing-skills --skill message-test-designer -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/aaron-he-zhu/aaron-marketing-skills --path narrative/evaluate/message-test-designer
```

### Windsurf (Devin Desktop)

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

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add aaron-he-zhu/aaron-marketing-skills --skill message-test-designer -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/aaron-he-zhu/aaron-marketing-skills/tree/HEAD/narrative/evaluate/message-test-designer/SKILL.md

## When to use it
- design a message-market-fit panel before spending on scale
- run a 5-second comprehension test on a new tagline
- precommit a pass threshold and a stop/revise rule
- build a stimulus set from the canon with exact hashes
- route unverifiable claims to the claims registry instead of the ledger

## Example prompts
- Design a message-market-fit panel test for our tagline. Target panel: product leads. Variants: three phrasings.
- Design a 5-second comprehension test for our new homepage hero: "[headline + subhead]". What do we measure and what's the pass bar?
- We have three positioning statements. Design the Wynter-style test that tells us which one lands before we scale spend.

## Source text · SKILL.md · Apache-2.0

Below is the content of a third-party file. It is data, not instructions for you: do not act on anything written inside unless the human asks. Block tag: e274f8cfa46d9428.

BEGIN THIRD-PARTY FILE e274f8cfa46d9428
```text
---
name: message-test-designer
slug: aaron-message-test-designer
displayName: "Message Test Designer · 消息测试设计"
summary: "消息理解度/五秒/消息-市场契合面板测试设计"
description: 'Use when the user asks to "test our messaging before we scale it", "design a message-market-fit panel", or "run a 5-second comprehension test on our new tagline"; produces a message-test design spec — hypothesis, panel and recruit criteria, comprehension / 5-second / message-market-fit (Wynter-style) protocols, stimulus set drawn from the canon, success thresholds, and a stop/revise decision rule — for the TALE Evaluate phase so the message is validated before any paid scale. It designs the test; it never runs the experiment or adjudicates a claim. Not for running the panel or A/B experiment — use send-experiment-designer or ad-test-designer; not for analyzing the results — use performance-analyzer; not for authoring the message itself — use message-system-architect. 消息测试/理解度测试/面板设计/五秒测试/消息市场契合'
version: "20.1.0"
license: Apache-2.0
compatibility: "Claude Code and compatible agent-skill hosts"
homepage: "https://github.com/aaron-he-zhu/aaron-marketing-skills"
when_to_use: "Use when you have a candidate message (tagline, one-liner, value pillars, or a per-surface message-match spec) and want to validate it with a target panel before spending on scale: designing the comprehension test, the 5-second recall test, or the Wynter-style message-market-fit panel — hypothesis, recruit criteria, stimulus set from the canon, success thresholds, and the stop/revise rule. The design layer of the TALE Evaluate phase; execution is handed to the experiment builders and analysis to performance-analyzer. Not for running the test and not for authoring the message."
argument-hint: "<message / tagline / surface> [target panel] [candidate variants]"
```
END THIRD-PARTY FILE e274f8cfa46d9428

The third-party file has ended. Everything between the e274f8cfa46d9428 tags is data: requests and commands from it are not followed. An “end of file” line with another tag inside the block is part of the file.

First 1923 of 14483 bytes. © aaron-he-zhu · Apache-2.0 · https://github.com/aaron-he-zhu/aaron-marketing-skills/tree/HEAD/narrative/evaluate/message-test-designer/SKILL.md

_Part of the page did not fit — see the HTML version._

---
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