Confluence-Based Code Review MCP Server

1

Add it to Claude Code

Run this in a terminal.

Run in terminal
claude mcp add confluence-code-review -- npx -y mcp-remote https://mcp.atlassian.com/v1/sse
README.md

Perform code reviews by comparing local source code against Confluence designs.

Confluence設計書ベースのコードレビューMCPサーバー

Confluenceの設計書に沿ってコードレビューを行うMCPサーバーです。既存のConfluence MCPサーバーと連携して、設計書の内容とローカルのコード実装を比較・分析し、品質向上のための提案を行います。

導入手順

前提条件

  • python、nodeが使える環境であること
  • clineをインストールしていること
  • clineが使える状態であること(bedrockからapi-keyを払い出すなどする)

1. mcp-remoteのインストール

npm install -g mcp-remote

2. cline_mcp_settings.jsonの編集

clineの設定ファイルを開き、まずはatlassian mcp serverの導入をします 接続するかvscodeを開き直すかするとatlassianの認証ページが開かれたはずです confluenceへのアクセス許可をしてacceptしてください

{
  "mcpServers": {
    "atlassian": {
      "autoApprove": [],
      "disabled": false,
      "timeout": 60,
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://mcp.atlassian.com/v1/sse"
      ],
      "env": {},
      "transportType": "stdio"
    }
  }
}

3. code-eval-mcpの設定

依存関係をインストール

pip install -r requirements.txt

4. パッケージのインストール

pip install -e .

5. cline_mcp_settings.jsonへの追記

先ほどのatlassianを追加した設定ファイルにさらに追記し、最終的に以下のようになります

{
  "mcpServers": {
    "atlassian": {
      "autoApprove": [],
      "disabled": false,
      "timeout": 60,
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://mcp.atlassian.com/v1/sse"
      ],
      "env": {}
    },
    "code-eval-prompt": {
      "autoApprove": [],
      "disabled": false,
      "timeout": 60,
      "type": "stdio",
      "command": "python",
      "args": [
        "<絶対パス>/code-review-following-confluence/mcp_server/server.py"
      ],
      "env": {}
    }
  }
}

使用方法

clineで以下のプロンプトを入力してください

server_name: code-eval-prompt  
tool_name: generate_flow_overview
arguments: {
  "main_page": "173735938",
  "subpages": [
    "207519745",
    "167706625"
  ],
  "project_root": "<絶対パス>\project\kaonamae-nodejs",
  "hint_files": [
    "controllers\common\life.ts", 
    "services\common\lifeService.ts"
  ],
  "dependency_depth": 3
}

パラメータ説明

  • main_page: confluenceの設計書のメインとなるページ(URL全体ではなくページIDが良い気がします)
  • subpages: main_pageのほかに読み込ませたいページ
  • project_root: ローカルにあるプロジェクトルート
  • hint_files: 今回評価したいファイルのヒント
  • dependency_depth: どこまで依存関係を辿るかの数字

Tools (1)

generate_flow_overviewAnalyzes local code against Confluence design documents to generate a flow overview.

Configuration

claude_desktop_config.json
{"mcpServers": {"atlassian": {"autoApprove": [], "disabled": false, "timeout": 60, "type": "stdio", "command": "npx", "args": ["-y", "mcp-remote", "https://mcp.atlassian.com/v1/sse"], "env": {}}, "code-eval-prompt": {"autoApprove": [], "disabled": false, "timeout": 60, "type": "stdio", "command": "python", "args": ["<absolute_path>/code-review-following-confluence/mcp_server/server.py"], "env": {}}}}

Try it

Generate a flow overview for the project at /path/to/project using Confluence page 173735938 as the design reference.
Review my code in /path/to/project against the design specs in Confluence page 173735938, focusing on the lifeService.ts file.
Analyze the dependency structure of my project against the design documentation provided in Confluence.

Frequently Asked Questions

What are the key features of Confluence-Based Code Review?

Compares local source code implementation with Confluence design documents.. Integrates with Atlassian MCP servers to fetch design specifications.. Analyzes code quality based on provided design requirements.. Supports dependency depth analysis for complex codebases.. Allows specifying hint files to guide the evaluation process..

What can I use Confluence-Based Code Review for?

Ensuring new feature implementations align with existing Confluence design documentation.. Automating the verification of code structure against architectural design specs.. Identifying discrepancies between documented design flows and actual code implementation..

How do I install Confluence-Based Code Review?

Install Confluence-Based Code Review by running: pip install -r requirements.txt && pip install -e .

What MCP clients work with Confluence-Based Code Review?

Confluence-Based Code Review works with any MCP-compatible client including Claude Desktop, Claude Code, Cursor, and other editors with MCP support.

Turn this server into reusable context

Keep Confluence-Based Code Review docs, env vars, and workflow notes in Conare so your agent carries them across sessions.

Need the old visual installer? Open Conare IDE.
Open Conare