Shelpuk & Company
    Open Source
    MCP Server

    Lad Code Review

    Project-aware AI code review for AI coding agents. The only AI code review MCP that understands your project context. Dual-reviewer architecture catches what single-model reviews miss.

    View on GitHub

    "Fantastic setup. Improved my Claude Code generation quality by 15–20% for our company’s commercial projects."

    Ostap Dribniuk
    Ostap DribniukSenior Software Engineer at Relevant Software

    Part of Our AI Coding Suite

    TDD for AI Agents orchestrates our complete open-source toolkit. It calls Kindly to verify assumptions and Lad to review designs and code — ensuring all three tools work together seamlessly.

    Kindly Web SearchLad Code ReviewTDD Skill

    Why Lad is Different

    Zero-Config OpenRouter

    Fetches model metadata directly from OpenRouter API. New models just work—no manual configuration needed.

    Dual-Reviewer Mode

    Two reviewers run in parallel (Kimi K2 + GLM-4.7 by default). Reduces bias and catches more issues.

    Project-Aware Review

    Optional Serena integration provides long-term project memory, codebase indexing, and historical context.

    Workflow Integration

    Separate tools for system design review and code review. Both support text input and file references.

    MCP Tools

    Each tool runs two OpenRouter-backed reviewers in parallel and returns both outputs plus a synthesized summary.

    system_design_review

    Reviews architectural proposals, design documents, and planning decisions.

    code_review

    Reviews implementation code, diffs, and file changes.

    Default Reviewer Models

    Primary: moonshotai/kimi-k2-thinking
    Secondary: z-ai/glm-5-turbo

    Both are current SOTA open-source models. Configurable via environment variables.

    What Lad Replaces

    Manual OpenRouter model configuration
    Context-blind AI code reviewers
    Separate tools for design vs. code review
    Constantly re-explaining requirements

    One Suite, Three Tools

    TDD for AI Agents is the orchestrator that ties our open-source AI coding tools into a complete development workflow. Each tool has a specific role.

    Kindly Web Search

    Verifies API docs, finds solutions, retrieves full technical context in a single call.

    Lad Code Review

    Provides a fresh pair of AI eyes — catches coherence-driven mistakes the coding agent misses.

    TDD Skill (Orchestrator)

    Enforces the workflow, writes requirements, coordinates research and review steps.

    Supported Clients

    Lad works with any MCP client that supports stdio servers:

    Codex (OpenAI)
    Claude Code
    Gemini CLI
    Cursor
    Windsurf
    Antigravity (Google IDE)

    Ready to catch what your AI coding agent misses?

    Deploy Lad in your private cloud or hire the engineers who built it.

    Star on GitHub

    Frequently Asked Questions

    Answers to common questions about this topic.

    How can we improve the quality of AI-assisted code review?

    AI-powered code review tool that enforces best practices, catches bugs early, and improves code quality across your team. Shelpuk AI Technology Consulting designed Lad from hands-on experience reviewing AI-generated code and architecture proposals in delivery programs.

    How does Lad's dual-reviewer approach improve code and system-design decision quality?

    Lad combines multiple reviewer perspectives to surface risks, contradictions, and missing assumptions that single-pass reviews often miss. Shelpuk AI Technology Consulting uses this framework to help teams increase review signal quality and reduce avoidable regression risk.

    What is the recommended integration approach for Lad in engineering pipelines?

    The strongest setup links Lad checks to pull-request and architecture review steps with explicit criteria for severity, confidence, and follow-up. Shelpuk AI Technology Consulting supports implementation in MCP-compatible client environments with practical rollout and operating guidance.

    How should teams think about cost when adopting open-source Lad in production?

    The tool itself is open source, while production adoption cost usually comes from workflow integration, policy tuning, and reviewer quality governance. Shelpuk AI Technology Consulting provides advisory and delivery services for organizations that need reliable quality gates around AI coding agents.