This skill structures product discovery with Teresa Torres’ Opportunity Solution Tree. It starts from a single measurable outcome, maps three to seven customer opportunities from the research provided, ranks them with an Opportunity Score, generates at least three solutions for each top opportunity and designs experiments to test the best ones.
The tree is presented as a clear hierarchy and saved as markdown when it is large.
What You Get
- A four-level tree: outcome, opportunities, solutions and experiments
- Opportunities ranked by Opportunity Score or a qualitative assessment
- At least three solution ideas for each prioritized opportunity
- One or two experiments per promising solution, each with a success threshold
Common Use Cases
- Turning interview and survey findings into a map of customer needs
- Choosing what to build next for a retention or activation goal
- Reviewing a roadmap that jumps straight to solutions
What Makes It Unique
It carries the principles of continuous discovery in the prompt: one outcome at a time, opportunities and not features, at least three solutions before choosing, and weekly updates. The README lists it as a standalone reference that Claude draws on whenever relevant, so no command is needed.
Under The Hood
The skill is a prompt only, with no scripts. It tells the agent to use the outcome and any research data the user provides, such as interviews, surveys or analytics exports, and to save the tree as markdown if it is large. It makes no web or API calls and runs no commands.