# Small product assumption test

A found skill by terminal-skills.

Turns a product idea into a list of assumptions, then designs the cheapest tests and interview questions to check the riskiest ones first.

Task: Planning and reviews
Works with: Claude, ChatGPT, Gemini
Tools: Claude
Difficulty: Intermediate
Country: Country unknown

Use this when you need to decide what to validate next on a product, starting with the cheapest evidence rather than opinions.
Skip this if you want a full business-idea go or no-go, or if you only want a past-behaviour interview script.

Credit: terminal-skills
Source: https://github.com/TerminalSkills/skills/blob/main/skills/product-discovery/SKILL.md
Licence: Apache License 2.0

## Skill text

Please turn the text below into a reusable skill. If you are Claude, save it as a skill. If you are ChatGPT or Gemini, save it as the instructions for a Project, custom GPT or Gem. If you cannot save skills, just follow these instructions now.

Name: Small product assumption test

Description: Use this when you need to decide what to validate next on a product, starting with the cheapest evidence rather than opinions.

Ask me for:
1. The product or change I am considering, and who it is for.
2. The outcome I want, in one line.
3. Any evidence I already have, if any.

Steps:
1. Restate the desired outcome in one line.
2. List the main assumptions underneath it, grouped as demand, value, usability or feasibility.
3. Rank them by how much damage they would do if wrong, and how little evidence I have.
4. For the top assumptions, suggest the cheapest test first, such as a conversation, a simple landing test or a manual concierge run, and say what result would confirm or kill the assumption.
5. Draft a short interview guide that asks about past behaviour, not whether someone would use the idea.
6. These are published discovery practices, including work associated with Teresa Torres and Alberto Savoia. Use them in your own words. Do not speak as those authors, quote them, or claim they endorse this skill.

Output: The outcome line, the ranked assumptions, the test plan and the interview guide.

Rules:
- Do not invent interview findings or test results.
- Do not tell me to spend money or launch a test without my yes.
- If I lack a clear outcome, ask me before going further.

This is our own short starter, informed by the product-discovery skill from TerminalSkills. The original is at https://github.com/TerminalSkills/skills/blob/main/skills/product-discovery/SKILL.md. Apache License 2.0, skill metadata author terminal-skills. No separate NOTICE file was found. Check the licence before reusing its own text. No full third-party text copied.

## More

- [This skill as a web page](https://agentstories.org/skills/small-product-assumption-test.html)
- [All skills, as a Markdown list](https://agentstories.org/skills.md)
- [Skills directory](https://agentstories.org/skills/)
