MCP.so
登录

Mcpn MCP Server

@dx-zero

基本信息

分类

其他

许可证

MIT

传输方式

stdio

作者

dx-zero

关于 Mcpn MCP Server

orchestrate & combine prompts + MCP servers into compound MCP tools

Mcpn MCP Server 配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "mcpn": {
      "command": "npx",
      "args": [
        "-y",
        "mcpn@latest",
        "init"
      ]
    }
  }
}

Mcpn MCP Server 工具

2

Reflect on a thought and produce a reflection/new set of thoughts

Deeply reflect upon the provided idea/problem/question

概览

What is Mcpn?

Mcpn (mcpn.ai) is an MCP server that helps you create, organize, and share AI workflows and prompts as reusable, version‑controlled YAML configurations across MCP tools and servers.

How to use Mcpn?

Install via npx -y mcpn@latest init or add a JSON config to your MCP client (e.g., Cursor). Optionally supply --config /path/to/.workflows and --preset thinking,coding,github to load custom or preset workflow definitions.

Key features of Mcpn

  • Define workflows in YAML files (.workflows or .mcp-workflows)
  • Combine prompts with MCP tools in reusable templates
  • Inject typed parameters using {{ parameter_name }}
  • Choose between situational (default) and sequential tool execution
  • Reduce token usage by routing prompts through a dedicated tool
  • Version control workflows with git for team sharing

Use cases of Mcpn

  • Debug web applications by invoking browser logs and screenshots in sequence
  • Perform structured PR reviews and PR creation using GitHub CLI
  • Plan code changes with systematic analysis and implementation steps
  • Think through complex problems with multi‑perspective reasoning modes
  • Automate incident response by gathering logs, checking issues, and writing reports

常见问题

1

What dependencies does Mcpn require?

Mcpn is an npm package run via `npx`, so Node.js must be installed on your system.

2

How can I share workflows with my team?

Store your YAML files in a `.workflows` or `.mcp-workflows` directory and commit them to a git repository. Team members then point to the same config folder using `--config`.

3

Which presets are included out of the box?

Three presets are available: `thinking` (reasoning modes), `coding` (debugging, architecture, planning, PRD), and `github` (PR review, PR creation, branch creation, save changes).

4

How do I define custom workflows with parameters?

Create a YAML file with a workflow name, description, prompt containing `{{ parameter_name }}`, and a `parameters` block specifying each parameter’s type, description, and required flag.

5

Can I control the order tools are used in a workflow?

Yes. Set `toolMode: sequential` to execute tools in the exact order listed; omit it (or set `situational`) to let the AI decide which tools to use as needed.

评论

其他 分类下的更多 MCP 服务器