sdd-review

Orchestrate design, implementation, and dead-code reviews with configurable engines.

Updated Jan 19, 2026
One-click install
npx skills add https://github.com/sync-dev-org/sync-sdd --skill sdd-review-sync-dev-org
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sdd-review
Source: https://github.com/sync-dev-org/sync-sdd/tree/main/framework/claude/skills/sdd-review
Command: npx skills add https://github.com/sync-dev-org/sync-sdd --skill sdd-review-sync-dev-org

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill streamlines the review process for code design, implementation, and dead code, ensuring consistent quality and adherence to specifications across different review types.

Core Features & Use Cases

  • Unified Review Types: Handles design, implementation, and dead-code reviews within a single, consistent pipeline.
  • Configurable Engines: Supports default SubAgent (Claude Code Agent) or external engines like Codex, Claude, and Gemini, with configurable effort levels.
  • Dynamic Inspector Generation: Leverages a Briefer to generate dynamic inspectors based on the review context, enhancing review thoroughness.
  • Use Case: Automatically trigger an implementation review for a new feature, with dynamic inspectors identifying potential issues specific to the changes, followed by an auditor's synthesis of the findings.

Quick Start

Initiate a design review for the 'user-authentication' feature by running the sdd-review skill with the argument 'design user-authentication'.

Frequently Asked Questions about sdd-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a unified code review pipeline for design and implementation?

A unified code review pipeline orchestrates design, implementation, and dead-code analysis by integrating multiple inspectors and an auditor. You can initiate a review by triggering the pipeline with a specific scope argument like 'design user-authentication'.

What is dynamic inspector generation in a code review process?

Dynamic inspector generation leverages a Briefer to create context-specific inspectors during a code review. This enhances review thoroughness by automatically identifying potential implementation issues specific to the changes being audited.

Can I use external engines like Codex or Gemini for automated code auditing?

Yes, automated code auditing supports configurable engines including SubAgent, Codex, Claude, and Gemini. You can adjust configurable effort levels to dynamically generate inspectors and synthesize findings across different review scopes.

What review scopes are supported for cross-checking and dead code analysis?

Dead code analysis and cross-checking support various review scopes including per-feature, wave-scoped, and cross-cutting concerns. The pipeline persists verdicts and archives review artifacts for each selected scope.

Does the implementation review pipeline work without external dependencies?

The implementation review pipeline operates with zero dependencies by defaulting to the SubAgent engine. It handles cross-cutting concerns and archives artifacts without requiring external tools or modules.