Angular PR Review MCP
Enables automated pull-request review workflows for Angular repositories by providing tools to generate, read, and save diffs, and to fetch review templates.
README
Angular PR Review MCP
Reusable MCP server for automated pull-request review workflows across multiple Angular repositories.
This package exposes three MCP tools:
get_pr_review_templategenerate_pr_diffread_pr_diffsave_pr_review
It runs in the target repository context (the current working directory), so one published package can be reused by any Angular app.
Requirements
- Node.js 20+
- Git installed and available in PATH
- VS Code with Copilot Chat MCP support
Local Development
npm install
npm run build
npm run dev
Publish To npm
If publishing publicly:
npm login
npm publish --access public
If publishing under a scoped package (recommended for account/org ownership):
- Change the
namefield in package.json to your scope, for example@your-scope/angular-pr-review-mcp. - Publish:
npm login
npm publish --access public
Create Separate GitHub Repository
From this folder, create and push to a dedicated repository in your account:
git init
git branch -M main
git remote add origin https://github.com/<your-username>/angular-pr-review-mcp.git
git add .
git commit -m "Initial reusable Angular PR review MCP package"
git push -u origin main
Use From Any Angular Repository
Add or update .vscode/mcp.json in each Angular repository:
{
"servers": {
"angular-pr-review": {
"type": "stdio",
"command": "npx",
"args": ["-y", "angular-pr-review-mcp"],
"cwd": "${workspaceFolder}",
"env": {
"PR_REVIEW_BASE_BRANCH": "main",
"PR_REVIEW_OUTPUT_DIR": "pr_review"
}
}
}
}
For a scoped package, replace angular-pr-review-mcp with your published name.
Environment Variables
PR_REVIEW_BASE_BRANCH: target base branch for comparison (defaultmain)PR_REVIEW_OUTPUT_DIR: review artifacts directory (defaultpr_reviewunder repo root)PR_REVIEW_REPO_ROOT: override repository root (default current working directory)PR_REVIEW_SERVER_NAME: MCP server name (defaultangular-pr-review)
Tool Workflow
generate_pr_diffwith commit SHA.read_pr_diffto retrieve diff content.save_pr_reviewto persist markdown review comments.
Recommended flow:
get_pr_review_templatefirst, to load the local review rubric and avoid re-deriving rules each run.generate_pr_diffandread_pr_difffor evidence collection.save_pr_reviewfor the final report.
Generated files are written into the configured output folder in each consuming repository.
Local Template Artifact
The repository includes pr_review_prompt_template.md so teams can version review guidance in source control.
The get_pr_review_template tool returns the same policy text (with optional commit token replacement), which helps keep reviews deterministic and reduces token consumption.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。