Backend Code Review Agent

This repository provides a prompt that configures an AI agent to act as an automated backend code reviewer, helping to boost productivity, improve code quality, and enhance the overall developer experience.

Features

✨ Code Review

Prompt directs an AI agent to perform an automated pre-review of code changes. It identifies potential issues and saves valuable time for human reviewers.

• 🔌 Requirements: Redmine MCP Server (tool: get_issue)
• 📄 Output: A structured report of findings using Conventional Comments.
• 🚧 TODO: Integrate checks against a specific backend code style guide

Getting Started

Follow these steps to set up and use the prompt in your development workflow.

Prerequisites

• Node.js and npm installed
• Client supporting MCP Servers (See: MCP - Example Clients)

1. Clone this repository

Clone this repository to get local access to the prompt.

2. Setup required MCP Servers

• Redmine:
  • Clone Redmine MCP Server (GitHub repository)
  • Run npm install
  • Run npm run build

3. Configure new MCP Server in your IDE

Add MCP Server configuration to the configuration file. Cursor path is ~/.cursor/mcp.json or could be opened via UI in Cursor Settings -> Tools & MCP -> New MCP Server

• Redmine: (Cursor)

{
  "redmine": {
    "type": "stdio",
    "command": "node",
    "args": ["<path-to-cloned-mcp-server-repo>/dist/index.js"],
    "env": {
      "REDMINE_BASE_URL": "https://<your_redmine_domain>/",
      "REDMINE_API_KEY": "<your_redmine_api_key>"
    }
  }
}

Enable the tools the prompt requires for the configured MCP Server. Example: get_issue for the Code Review prompt.

4. Use the prompt (Cursor)

• Open AI Agent chat panel
• Reference the local prompt file using @, for example: @code_review.md
• The agent will then execute the steps defined in the prompt and provide the output

License

This project is licensed under MIT.