Rules MCP Server
An MCP server that allows coding agents to look up contextual rules and patterns on demand, providing just-in-time guidance for specific tasks like writing tests or authoring UI.
Tools
get_rules_for_writing_tests
When writing tests, ALWAYS ask this tool for rules on how to write the test first. Then, use the rules to write the test.
README
My Rules
An MCP server for all my rules, prompts, etc etc. Allows agents to call rules on demand.
There's no point filling 4 paragraphs of instructions for how to write tests if your current session isn't going to write a test.
Spiritually similar to Cursor's rules.
Usage
Claude Code
claude mcp add --transport http rules https://mcp.cianfrani.dev/mcp
Cursor
Claude Desktop
{
"rules": {
"command": "npx",
"args": [
"mcp-remote",
"https://mcp.cianfrani.dev/mcp"
]
}
}
What's It Do?
Allows coding agents to lookup rules on demand within the context of their current task.
> write a test for @calc.py
⎿ Read calc.py (26 lines)
⏺ rules:get_rules_for_writing_tests (MCP)(language: "python")
⎿ - Before testing a protected method, ask "Can this logic be adequately tested through the public interface?". If yes, don't test the protected method directly.
If no, continue testing the protected method.
- Remember to write tests using the Arrange, Act, Assert pattern.
… +12 lines (ctrl+r to expand)
⏺ Write(test_calc.py)
Why?
- It's really annoying trying to carry all these little notes between projects/IDEs.
- I don't want to have to manually invoke rules.
How's It Work?
Clearly-defined tool descriptions allow the agent to fetch rules on demand, only if they appear to be relevant.
What Doesn't work?
- Patterns must be associated with a specific action. For example, "writing tests", "authoring UI", "reviewing a PR".
- The agent ultimately decides if it's going to call the tool. Sometimes it does. Sometimes it doesn't.
Resources
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。