Translate the idea into a buildable and testable specification covering requirements, data models, user flows and acceptance criteria.
Learning outcome
An MVP specification clear enough to hand to a development agent.
Use fictional or sanitised examples. Never paste passwords, API keys, confidential documents or sensitive personal data into a public AI service.
Workflow
- Freeze the MVP purpose and scope.
- Write user stories and acceptance criteria.
- Define data, rules and error states.
- Check that every requirement is testable.
Master Prompt Template
Act as a business analyst and Python application architect.
Use the following validated idea:
[PASTE MODULE 1 OUTPUT]
Prepare an MVP specification with these sections:
1. Purpose, users and business outcome.
2. Scope and out-of-scope items.
3. Actors, roles and access permissions.
4. User stories prioritised as Must/Should/Could.
5. Functional requirements with unique IDs FR-001 onwards.
6. Non-functional requirements: performance, accessibility, privacy, security and maintainability.
7. Primary flows, empty states, errors and recovery.
8. Data entities, fields, types, relationships and validation rules.
9. The simplest suitable Python stack and selection rationale.
10. Given/When/Then acceptance criteria for every major function.
11. Risks, assumptions, dependencies and open questions.
Prefer a simple, testable solution. Do not add out-of-scope features without labelling them as recommendations.Practical exercise
Ask a user to read the specification without a verbal explanation. Record unclear sections and ensure every major function has acceptance criteria.
Completion checklist
- Every requirement has an ID.
- Roles and access are stated.
- The data model has validation rules.
- Acceptance criteria are testable.