What categories of questions does the agent ask?

The agent focuses on six key areas: Design, Edge Cases, Clarity, Philosophy, Failure Modes, and Testing.

What depth levels are available?

You can choose from 'Quick' (3-5 questions), 'Standard' (7-10 questions), or 'Deep' (15-20 questions) depending on the complexity of the code.

How does Socratic Review differ from a standard code review?

Traditional reviews provide direct feedback and fixes, whereas Socratic Review uses probing questions to help you discover issues and articulate your own reasoning.

Can I use this skill in my CI/CD pipeline?

Yes, the skill includes a non-interactive mode that automatically synthesizes questions and insights into structured JSON for automated PR comments.

Does it automatically fix my code?

No, its primary purpose is learning and insight. It provides recommendations at the end, but the goal is for the developer to own the implementation and logic.

Socratic Code Review

Name: Socratic Code Review
Author: rysweet

byrysweet

•

Collaboration Tools

Facilitates deeper code understanding through interactive, question-based dialogue rather than direct criticism.

About

Socratic Code Review is a specialized skill for Claude Code that transforms the standard review process into a collaborative mentorship session. Based on the Feynman Technique, it uses probing questions about design choices, edge cases, and architectural philosophy to help developers articulate their reasoning and uncover hidden assumptions. This approach is specifically designed for scenarios where developer growth, code ownership, and long-term maintainability are more critical than immediate, superficial fixes.

Key Features

Interactive Q&A covering design, edge cases, and failure modes
Automated non-interactive mode for CI/CD integration and PR commenting
Direct persistence of design insights to project documentation files
Adjustable review depths ranging from quick checks to deep architectural dives
Final synthesis reports identifying surfaced assumptions and specific recommendations
16 GitHub stars

Use Cases

Onboarding new developers by encouraging them to explain their implementation choices
Reviewing complex or high-risk infrastructure where understanding the 'why' is critical
Formalizing architectural decision records through dialogue-driven insights

About

Key Features

Interactive Q&A covering design, edge cases, and failure modes
Automated non-interactive mode for CI/CD integration and PR commenting
Direct persistence of design insights to project documentation files
Adjustable review depths ranging from quick checks to deep architectural dives
Final synthesis reports identifying surfaced assumptions and specific recommendations
16 GitHub stars

Use Cases

Onboarding new developers by encouraging them to explain their implementation choices
Reviewing complex or high-risk infrastructure where understanding the 'why' is critical
Formalizing architectural decision records through dialogue-driven insights