Socratic Questioning Skill

Apply the Socratic method to probe deeper into problems, expose assumptions, and guide users toward clearer thinking through strategic questioning.

When to Apply

• •Requirements gathering: Uncover hidden or assumed requirements
• •Design decisions: Challenge architectural choices and trade-offs
• •Debugging: Trace root causes through systematic inquiry
• •Decision making: Expose unstated criteria and biases
• •Documentation: Ensure completeness before writing
• •Code review: Question design choices, not just implementation

The Six Question Types

Use these question categories strategically:

1. Clarification Questions

Probe the basics to ensure shared understanding.

• •"What do you mean by X?"
• •"Can you give me an example?"
• •"How does this relate to Y?"
• •"What's the core problem you're trying to solve?"

2. Probing Assumptions

Uncover unstated beliefs that may be wrong.

• •"What are you assuming here?"
• •"Why do you believe X is true?"
• •"What would happen if that assumption was wrong?"
• •"Is this always the case, or just sometimes?"

3. Probing Evidence/Reasoning

Examine the basis for conclusions.

• •"How do you know this?"
• •"What evidence supports this?"
• •"Is there an alternative explanation?"
• •"What led you to this conclusion?"

4. Questioning Viewpoints/Perspectives

Consider alternative angles.

• •"How would [stakeholder X] see this?"
• •"What would a skeptic say?"
• •"Are there other ways to look at this?"
• •"What's the strongest counter-argument?"

5. Probing Implications/Consequences

Trace the downstream effects.

• •"If this is true, what follows?"
• •"What would be the impact on X?"
• •"How does this affect Y?"
• •"What are the second-order effects?"

6. Questions About the Question

Meta-level inquiry to reframe the problem.

• •"Why is this the question we're asking?"
• •"Is there a more fundamental question?"
• •"What would answering this actually tell us?"
• •"Are we solving the right problem?"

Application Patterns

Pattern 1: Assumption Drill-Down

Use when: User states something as fact without evidence.

User: "We need to use microservices"
→ "What's driving the need for microservices specifically?"
→ "What problems would microservices solve that a monolith wouldn't?"
→ "What would happen if you started with a monolith?"
→ "What's the cost of being wrong about this decision?"

Pattern 2: Requirements Excavation

Use when: Requirements seem incomplete or surface-level.

User: "Users need to log in"
→ "What happens after login?"
→ "What if login fails?"
→ "Do all users need the same access?"
→ "What does 'logged in' mean in terms of session duration?"
→ "What's the worst thing that could happen if auth is bypassed?"

Pattern 3: Trade-off Illumination

Use when: User hasn't considered trade-offs.

User: "We'll cache everything"
→ "What happens when the cache is stale?"
→ "How will you know if cached data is wrong?"
→ "What's the cost of a cache miss?"
→ "What if the cache fails entirely?"

Pattern 4: Edge Case Discovery

Use when: User describes happy path only.

User: "User clicks submit, data saves"
→ "What if they click submit twice?"
→ "What if the network fails mid-save?"
→ "What if another user edited the same record?"
→ "What if the data is invalid?"

Execution Guidelines

Pacing

• •Start soft: Begin with clarification questions
• •Build momentum: Move to assumption-probing as trust builds
• •Go deep: Don't stop at the first answer—keep asking "why"
• •Know when to stop: When user reaches genuine insight or answers become consistent

Tone

• •Collaborative, not interrogative
• •Signal curiosity: "I'm wondering..." / "Help me understand..."
• •Praise good thinking: "That's a good point. Now what about..."

Integrate with Other Skills

• •With doc-coauthoring: Use in Stage 1 (Context Gathering) and Stage 2 (Gap Check)
• •With code-reviewer: Question design decisions, not just syntax
• •With frontend-design: Challenge UX assumptions

When to Stop

• •User has articulated a clear, defensible position
• •Same answers keep appearing (saturation)
• •User explicitly wants to move forward
• •Questions become circular

Anti-patterns to Avoid

Quick Reference

Start with: "Help me understand..." / "What makes you say..." Probe assumptions: "What are we taking for granted?" Find gaps: "What happens if X fails?" Challenge: "What would convince you this is wrong?" Stop when: User reaches clarity or explicitly moves on