Epic Scoping Skill

Interactive question bank-driven scoping for epic workflows.

Purpose

Systematically scope epics with structured AskUserQuestion calls, eliminating ambiguity before planning. Uses centralized question bank to ensure consistent, comprehensive scoping.

When to Invoke

•Always for new epics (via /epic command)
•Skip if resuming with /epic continue

Quick Start

bash

# Load question bank
cat .claude/skills/epic/references/question-bank.md

5-Round Interactive Scoping

Round 1: Initial Scoping (2 questions)

Questions from bank:

•initial_scoping.business_goal
•initial_scoping.subsystem_selection

javascript

AskUserQuestion({
  questions: [
    questionBank.initial_scoping.business_goal,
    questionBank.initial_scoping.subsystem_selection,
  ],
});

Apply answers to epic-spec.md:

•business_goal → ## Objective > ### Business Value
•subsystem_selection → ## Subsystems (update Involved field)

Round 2: Scope Refinement (0-4 questions, conditional)

Based on subsystem_selection, ask relevant follow-ups:

javascript

const refinementQuestions = [];

if (selectedSubsystems.includes("Backend API")) {
  refinementQuestions.push(questionBank.scope_refinement.backend_scope);
}
if (selectedSubsystems.includes("Frontend UI")) {
  refinementQuestions.push(questionBank.scope_refinement.frontend_scope);
}
if (selectedSubsystems.includes("Database")) {
  refinementQuestions.push(questionBank.scope_refinement.database_scope);
}
if (selectedSubsystems.includes("External integrations")) {
  refinementQuestions.push(questionBank.scope_refinement.integration_scope);
}

if (refinementQuestions.length > 0) {
  AskUserQuestion({ questions: refinementQuestions });
}

Apply answers:

•backend_scope → ### Backend section
•frontend_scope → ### Frontend section
•database_scope → ### Database section
•integration_scope → ### Infrastructure section

Round 3: Success Metrics (2 questions)

javascript

AskUserQuestion({
  questions: [
    questionBank.success_metrics.measurement_approach,
    questionBank.success_metrics.target_values,
  ],
});

// If "Specific targets" selected, follow-up:
if (answers["Targets"] === "Specific targets") {
  const customTargets = answers["Targets_custom"];
  // Parse and apply to epic-spec.md
}

Apply answers:

•measurement_approach → ### Success Metrics section
•target_values → ### Success Metrics (actual values)

Round 4: Dependencies & Constraints (2 questions)

javascript

AskUserQuestion({
  questions: [
    questionBank.dependencies_and_constraints.external_dependencies,
    questionBank.dependencies_and_constraints.constraints,
  ],
});

Apply answers:

•external_dependencies → ## Dependencies section
•constraints → ### Constraints section

Round 5: Complexity Assessment (2 questions)

javascript

AskUserQuestion({
  questions: [
    questionBank.complexity_assessment.technical_complexity,
    questionBank.complexity_assessment.sprint_estimate,
  ],
});

// Warn if 7+ sprints
if (answers["Sprint Estimate"] === "7+ sprints") {
  console.log("Warning: Consider breaking into smaller epics");
}

Apply answers:

•technical_complexity → YAML frontmatter complexity field
•sprint_estimate → ## Overview section

Progress Indicators

Show progress after each round:

code

🔍 Epic Scoping: Round 1/5 (Initial scoping)
📝 Applied: Business Goal → New capability
📝 Applied: Subsystems → Backend API, Frontend UI, Database
🔍 Epic Scoping: Round 2/5 (Scope refinement)
...

Completion Summary

After 5 rounds:

code

✅ Epic Scoping Complete (9 questions answered)

📊 Epic Summary:
  Business Goal: New capability
  Subsystems: Backend API, Frontend UI, Database (3)
  Success Metrics: User adoption (target: 80% within 30 days)
  Dependencies: Third-party APIs (Stripe)
  Constraints: Time constraint (2-week deadline)
  Complexity: Medium
  Estimated Sprints: 3-4 sprints

📋 Next Steps:
  1. Review epic-spec.md for accuracy
  2. Run /plan to generate research → plan → sprint breakdown
  3. Run /epic continue to execute automated workflow

Velocity Benefit

•With skill: 5-10 minute interactive scoping
•Without skill: 30+ minute unstructured discussion
•Improvement: 3-6x faster

Question Bank Reference

Location: .claude/skills/epic/references/question-bank.md

Contains:

•Initial scoping (2 questions)
•Scope refinement (0-4 conditional)
•Success metrics (2 questions)
•Dependencies & constraints (2 questions)
•Complexity assessment (2 questions)

Total: 8-9 questions across 5 rounds