discovery-interview
Conduct an interactive discovery interview to produce a structured product specification. Triggers: write a spec, PRD, feature spec, requirements, product requirements, scope a project, brainstorm a feature, flesh out an idea, plan a new project. Uses AskUserQuestion for all user choices; WebSearch/WebFetch when the user wants research. Outputs: user stories, acceptance criteria, technical constraints, prioritized requirements in docs/specs/ per SPEC_TEMPLATE.md. Do NOT use for: implementation, code review, debugging, refactors, or when the user already has a complete spec they only want edited.
95
100%
Does it follow best practices?
Impact
77%
1.08xAverage score across 3 eval scenarios
Advisory
Suggest reviewing before use
Evaluation results
59%
-13%Team Onboarding Portal Specification
Spec file location and template structure
Correct output directory
0%
0%
Date-prefixed filename
100%
100%
Executive Summary section
100%
0%
Problem Statement section
100%
37%
User Personas section
100%
100%
User Journey section
100%
50%
Functional Requirements with priority tiers
0%
30%
Technical Architecture section
100%
100%
Non-Functional Requirements section
100%
100%
Out of Scope section
100%
100%
Open Questions section
100%
75%
Research Findings appendix
0%
37%
100%
44%Payment Webhook Processing Service Specification
Technical architecture completeness
Data Model subsection
30%
100%
System Components subsection
70%
100%
Integrations subsection
30%
100%
Security Model subsection
30%
100%
Performance metric specificity
100%
100%
Scalability metric specificity
100%
100%
Reliability requirement
100%
100%
Security NFR entry
0%
100%
Auth mechanism in Security Model
41%
100%
72%
-15%Inventory Analytics Dashboard Specification
Research appendix and conflict resolution documentation
Research Findings appendix present
83%
0%
Research findings have specific content
100%
69%
Conflict resolution visible
100%
100%
Out of Scope section with explicit exclusions
100%
100%
Open Questions are specific
100%
100%
Success Criteria are measurable
100%
100%
Technology choices justified
100%
100%
All four NFR categories covered
100%
63%
Functional priority tiers present
0%
27%
- Repository
- kvokov/oh-my-ai
- Commit
29ba45a
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.