
Identify Assumptions (Existing Product)
Stress-test a feature idea for an existing product and surface its riskiest assumptions
- Repository
- phuryn/pm-skills
- Stars
- 27,616
- Licence
- MIT
- Files
- 1 file
What it does
When installed, the agent acts as a devil's advocate for a new feature idea you describe. It reviews the idea from product manager, designer, and engineer viewpoints, 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 current technology). For each assumption it notes what could go wrong, how confident it is, and a suggested way to test it. The goal is to strengthen the idea rather than reject it.
Great for
- Product managers preparing a feature idea for review or prioritisation
- Teams getting ready for an assumption mapping workshop
- Founders or leads who want a structured pre-mortem before committing engineering time
Skip it if
- Final decisions on their own: the output is AI-generated brainstorming and still needs validation with real users and data
- Products where the feature idea is not yet described; the skill relies on the user explaining the product, objective, market and feature
Try asking
- “Stress-test my idea for adding dark mode scheduling to our existing mobile app”
- “Identify risky assumptions for a new team-invite flow in our B2B SaaS dashboard”
- “Run a devil's advocate review of this PRD before our assumption mapping session”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 1 original files, the MIT licence and an attribution note.
identify-assumptions-existing.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip identify-assumptions-existing.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip identify-assumptions-existing.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Give the agent your product context, target market, objective and the feature idea up front, and attach any designs, PRDs or research files you have so it can read them first.
- Ask for the output in a format you can use next, such as a table of assumptions with confidence levels and tests, to feed into an assumption prioritisation canvas.
- Treat the High/Medium/Low confidence ratings as the agent's judgement, and validate the riskiest items with real customer research or experiments.
What's inside
- SKILL.md
Redistributed unmodified from phuryn/pm-skills under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.