product-discovery

Structures product discovery with opportunity-solution trees, assumption mapping, and experiment design.

111|73|Updated Mar 20, 2026
One-click install
npx skills add https://github.com/autopus-ai/autopus-adk --skill product-discovery-autopus-ai
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: product-discovery
Source: https://github.com/autopus-ai/autopus-adk/tree/main/.omp/skills/product-discovery
Command: npx skills add https://github.com/autopus-ai/autopus-adk --skill product-discovery-autopus-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Product teams often jump straight to building features without validating whether users actually need them, leading to wasted engineering effort. This Skill provides a structured discovery workflow that connects business outcomes to user opportunities, solution hypotheses, and low-cost validation experiments before any code is written. ## Core Features & Use Cases - Opportunity-Solution Tree (OST): Define measurable business and product outcomes, then map user unmet needs from feedback, support tickets, interviews, and funnel data into a prioritized opportunity table. - Assumption Mapping & Prioritization: Break each solution into value, usability, feasibility, and viability assumptions, ranked by impact and uncertainty so the riskiest assumptions get tested first. - Experiment Design: Choose the right validation method (Fake Door, Concierge, Wizard of Oz, Prototype, or MVP) with defined metrics, success criteria, and duration for each assumption. - Interview Script Generation: Produce non-leading interview scripts structured around warm-up, context, deep dive, and wrap-up phases, focused on past behavior rather than future intent. - Use Case: A PM notices high drop-off in onboarding. Use this Skill to map the underlying opportunities, generate three solution candidates, identify the riskiest value assumption, and design a Fake Door experiment with a measurable success threshold. ## Quick Start Use the product-discovery skill to run a discovery workflow for reducing onboarding drop-off, including an opportunity tree, key assumptions, and a validation experiment plan.

Frequently Asked Questions about product-discovery

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a product discovery process before building a feature?▼

Start by defining a measurable business and product outcome, then map user opportunities from feedback, support tickets, and funnel data into an opportunity-solution tree. Generate at least three solution candidates per opportunity, map their riskiest assumptions, and design a low-cost experiment to validate them before committing engineering resources.

What is an opportunity-solution tree and how do I build one?▼

An opportunity-solution tree connects a measurable outcome at the top to user opportunities (unmet needs) and then to solution candidates below. Build one by sourcing opportunities from interviews, support tickets, and usage data, scoring each by impact and confidence, then attaching multiple solution ideas to each opportunity.

Which experiment type should I use to test a product assumption?▼

Match the experiment to the assumption axis: Fake Door tests value quickly, Concierge validates value and usability manually, Wizard of Oz hides manual backend work to test feasibility, Prototypes test usability, and MVPs test everything at higher cost. Prioritize assumptions with high impact and high uncertainty first.

How do I write user interview questions that avoid bias?▼

Avoid leading questions and focus on past behavior rather than future intentions. Structure the script into warm-up, context, deep dive, and wrap-up phases, and use the 5 Whys technique to uncover root motivations behind stated problems.

When should I skip discovery and build directly?▼

Discovery adds the most value when uncertainty about user value or usability is high. For low-risk changes, well-understood problems, or urgent fixes, a full discovery cycle may cost more than it saves, so scale the effort to the assumption risk.