What are the technical requirements to use Code Lens?

To run Code Lens, you need Node.js version 24 or higher and a Google Gemini API key, which should be set as 'GEMINI_API_KEY' or 'GOOGLE_API_KEY'. It can be installed via npx or Docker and configured with various MCP-compatible clients.

Code Lens is a Model Context Protocol (MCP) server that utilizes Google Gemini for comprehensive AI-powered code analysis, automated pull request review, and documentation generation across your codebase.

How does Code Lens enhance the development workflow?

It provides an automated PR review pipeline with impact assessment, comprehensive file analysis for refactoring and code smell detection, logic verification using Gemini's sandbox, and natural-language Q&A for source files.

What unique features does Code Lens offer?

Code Lens integrates Google Gemini for advanced AI, Gemini's code execution sandbox for logic verification, structured and validated JSON outputs via Zod v4 schemas, and integrated web search with Google Grounding for up-to-date information.

Code Lens

Name: Code Lens
Author: j0hanz

byj0hanz

0•

安全与测试

开发者工具

数据科学与机器学习

Leverage Google Gemini for comprehensive code analysis, review, and automated documentation.

Code Lens is a Model Context Protocol (MCP) server that integrates Google Gemini's advanced AI capabilities directly into your development environment. It offers a powerful suite of tools for automating various aspects of the software development lifecycle, including in-depth code review, detailed file analysis, logic verification, and efficient documentation generation. By operating over standard I/O, Code Lens is designed for seamless integration with a wide range of IDEs and AI clients, providing intelligent assistance for tasks such as assessing pull request impact, identifying code smells, and generating refactoring suggestions.

主要功能

01Automated PR review pipeline with impact assessment and merge recommendations.

02Comprehensive file analysis for refactoring, code smell detection, and documentation generation.

03Logic verification of algorithms using Gemini's code execution sandbox.

04Structured and validated JSON outputs for all tools via Zod v4 schemas.

05Integrated web search with Google Grounding for up-to-date information retrieval.

060 GitHub stars

使用案例

01Conduct performance audits by analyzing time and space complexity of code changes.

02Automate pull request reviews, assess impact, and generate merge recommendations.

03Perform single-file analysis for refactoring suggestions, code smell detection, and documentation.