Guide · Agent skill
How to use the Identify Assumptions (Existing Product) agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
This guide explains how to install and use the identify-assumptions-existing agent skill, a set of instructions that makes an AI agent act as a devil's advocate for a new feature idea in an existing product. It is written for product managers, team leads and founders who want a structured review of risks before committing engineering time. By the end you will know what the skill does, how to add it to your agent, how to ask for it in plain language, and where its limits are.
Identify Assumptions (Existing Product)phuryn/pm-skills · MIT Download the skill Quick start
- 1Download the ZIP. Download the skill ZIP file from this page.
- 2Unzip into a skills folder. Unzip it into ~/.claude/skills/ for personal use, or into .claude/skills/ inside a project.
- 3Or upload in the Claude apps. In the Claude apps, go to Settings → Capabilities → Skills and upload the skill.
- 4Describe your feature idea. Explain the product, objective, market segment and feature idea in plain language.
- 5Ask for a risk assessment. Request a stress-test or risk assessment of the feature idea, or preparation for assumption mapping.
- 6Review the assumptions. Check each assumption's risk, confidence rating and suggested test.
- 7Validate with real data. Test the highest-impact, lowest-confidence assumptions with users and data before committing engineering time.
What does the Identify Assumptions skill make the agent do?
The skill is a single instruction file, SKILL.md, that tells the agent how to stress-test a feature idea. You describe the product, the objective, the market segment and the feature. The agent then reviews the idea from three viewpoints: a product manager (business viability, market fit, strategic alignment), a designer (usability, user experience, adoption barriers) and an engineer (technical feasibility, performance, integration challenges).
The agent then lists assumptions across four risk areas: Value (does it solve a real problem), Usability (can users learn it), Viability (can marketing, sales, finance and legal support it) and Feasibility (can it be built with existing technology, and are there integration risks). For each assumption it notes what could go wrong, how confident it is (High, Medium or Low) and a suggested way to test it. The skill states that the goal is to strengthen the idea, not reject it.
How do you install the skill?
Download the ZIP file for the skill from this page. Then install it using one of these two routes.
Option 1, personal folder: unzip the file into ~/.claude/skills/ so the skill is available across your projects.
Option 2, single project: unzip the file into the .claude/skills/ folder inside your project, so the skill is available only there.
Option 3, Claude apps: upload the skill in the Claude apps under Settings → Capabilities → Skills.
How do you trigger the skill with a natural request?
You do not need a special command. The skill's description covers stress-testing a feature idea, risk assessment and preparing for assumption mapping, so ask for one of those things in plain language. Describe the product, the objective, the market segment and the feature idea, since the skill relies on you explaining these.
If you have files such as designs, PRDs or research notes, attach or share them. The skill says to read any provided files first.
- Stress-test this feature idea for our project management app: a shared timeline view for enterprise customers.
- Do a risk assessment on our proposed in-app referral reward before we start building it.
- Prepare me for an assumption mapping workshop on a new onboarding checklist for our existing mobile app.
What should you put in your request?
The more context you give, the more specific the assumptions will be. The skill expects four things: the product, the objective, the market segment and the feature idea. Without a described feature idea the skill has nothing to work on.
- The product and what it does today
- The objective the feature is meant to serve
- The market segment or customers it is for
- The feature idea itself, in a few sentences
- Any designs, PRDs or research files you want the agent to read first
What does the output look like?
The output is a list of assumptions grouped by the four risk areas. Each entry explains what could go wrong, rates confidence as High, Medium or Low, and suggests a way to test the assumption. Treat the list as a starting point for discussion. Pick the lowest-confidence, highest-impact assumptions first and test them with real users and data.
What does the skill need, and is it safe to run?
The skill lists no requirements. It does not run scripts and does not use the network, according to its listing. It is a set of instructions for the agent to follow using the conversation you have with it. Check the skill's files yourself before installing if your organisation requires it. The listing names one file, SKILL.md.
- Requirements: none listed
- Runs scripts: no
- Uses network: no
- Files included: SKILL.md
What tips help you get better results?
Describe the feature concretely rather than as a vague theme. Say who it is for and what problem you believe it solves, because the Value assumptions depend on that. Ask for the assumptions to be turned into a table or a prioritised list if you plan to use them in a workshop. Finally, keep the output as a hypothesis list: the skill is brainstorming and its confidence ratings are not evidence.
What people use it for
- Stress-testing a new feature idea before it goes into a prioritisation review
- Preparing a list of assumptions to bring to an assumption mapping workshop
- Running a structured pre-mortem before a founder or lead commits engineering time
- Checking whether a feature will be usable and whether marketing, sales, finance and legal can support it
Frequently asked questions
How do I install the identify-assumptions-existing skill?
Download the ZIP from this page and unzip it into ~/.claude/skills/ for personal use, or into .claude/skills/ inside a project. You can also upload it in the Claude apps under Settings → Capabilities → Skills.
Is the identify-assumptions-existing skill free?
The skill is published under the MIT licence, according to its listing. The listing does not state any price, so check the repository for the licence terms that apply to you.
What do I need to run this skill?
The listing lists no requirements, does not run scripts and does not use the network. You do need an agent environment that can load skills, and you need to describe your feature idea.
Does the skill replace user research?
No. The output is AI-generated brainstorming and still needs validation with real users and data. Use it to decide which assumptions to test first.
What is the difference between Value, Usability, Viability and Feasibility?
Value asks whether the feature solves a real problem. Usability asks whether users can learn it. Viability asks whether marketing, sales, finance and legal can support it. Feasibility asks whether it can be built with existing technology.
Are there alternatives to this skill?
The source does not compare it with other tools, so this guide makes no comparison claims. The further reading on assumption prioritisation and risk management may help if you want to run the process by hand.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Use this skill if you are a product manager, team lead or founder who wants a structured pre-mortem of a feature idea before a review or workshop. Pick something else, or do the work with your own team and real users, if you need a final decision, if the feature is not yet described, or if you need evidence rather than brainstorming.
Download the skill