
适用于 pi 和模型上下文协议(MCP)服务器的单工具 MCPorter 桥扩展。
概览
该项目已不再维护。请使用 pi-mcp-client 以实现 Pi 中的原生 MCP 支持。 通过稳定的mcporter代理使用Pi中的MCP工具,并可选原生工具暴露,由MCPorter驱动。 - - 在有本地CLI时优先使用(如gh、git、kubectl、aws等)。- 当MCP能带来明确价值时,通过mcporter使用(例如:Linear、Slack、托管认证密集集成、跨工具工作流)。- 该软件包有意为之,不是取代pi正常的工具驱动工作流。 - 从一个稳定的mcporter代理开始,允许你选择选定的MCP工具进入更丰富的曝光度 - 使用你的MCPorter配置/运行时作为真实信息来源 - 支持发现(搜索)、模式帮助(描述)和执行(调用) - 返回关于认证/离线/http/stdio失败的有用错误提示 - Node.js 24 或更新版本 - MCPorter 至少安装并配置一台 MCP 服务器 3.
概述(AI 翻译)
[!警告] 该项目已不再维护。用途 pi-mcp-client 以实现原生 MCP Pi 中的支持。
通过稳定的 mcporter 代理使用 Pi 中的 MCP 工具,使用
可选的原生工具暴露,由
MCPorter。
🧠 理念:CLI > MCP
- pi 本质上是 CLI 优先的。
- 在有本地CLI时优先使用(
gh、git、kubectl、aws等)。 - 当MCP带来明确价值时,通过
mcporter使用(例如:线性、Slack、托管认证密集集成、跨工具工作流)。 - 该软件包有意作为一个薄桥接,而非替代Pi正常的工具驱动工作流。
✨ 为什么要用这个套餐
- 从一个稳定的
mcporter代理开始,允许你选择选定的 MCP 工具进入更丰富的曝光 - 使用你的 MCPorter 配置/运行时作为真实数据来源
- 支持发现(
search)、模式帮助(describe)和执行(call) - 返回关于认证/离线/http/stdio失败的有用错误提示
📋 先修条件
你需要:
- Node.js 24 或更新
- 安装并配置至少一台MCP服务器的MCPorter
🚀 安装
🚀 快速开始
-
确认MCPorter能看到你的服务器:
-
启动圆周率:
本页由 AI 翻译,技术细节请以原文为准。
README
以下为原文 README 节选,完整内容请查看原文。
🧳 pi-mcporter
[!WARNING] This project is no longer maintained. Use pi-mcp-client instead for native MCP support in Pi.
Use MCP tools from Pi through a stable mcporter proxy with
optional native tool exposure, powered by
MCPorter.
🧠 Philosophy: CLI > MCP
- pi is fundamentally CLI-first.
- Prefer native CLIs whenever available (
gh,git,kubectl,aws, etc.). - Use MCP via
mcporterwhen it adds clear value (for example: Linear, Slack, hosted auth-heavy integrations, cross-tool workflows). - This package is intentionally a thin bridge, not a replacement for pi’s normal tool-driven workflow.
✨ Why use this package
- Starts with one stable
mcporterproxy and lets you opt selected MCP tools into richer exposure - Uses your MCPorter config/runtime as source of truth
- Supports discovery (
search), schema help (describe), and execution (call) - Returns useful error hints for auth/offline/http/stdio failures
📋 Prerequisites
You need:
- Node.js 24 or newer
- MCPorter installed and configured with at least one MCP server
npm install -g mcporter
npx mcporter list # verify your servers are visible
🚀 Installation
pi install npm:pi-mcporter
🚀 Quick start
- Confirm MCPorter sees your servers:
npx mcporter list
- Start pi:
pi
- Ask for what you need and pi picks the right MCP tools automatically:
What are my open Linear issues this sprint?Catch me up on #engineering in Slack from today.Find the onboarding runbook in Notion and summarize the setup steps.
🔍 The three actions
The mcporter tool has three actions that map to a natural discovery → execution workflow.
search: find servers and tools
Use a known MCP server name or a natural-language capability phrase when you don’t know the exact tool selector:
{ "action": "search", "query": "linear", "limit": 5 }
{ "action": "search", "query": "find a tool for creating Linear issues", "limit": 5 }
Search returns matching server names and server.tool selectors with short descriptions. A server name confirms that MCPorter knows the integration, but it isn’t callable by itself. Choose a tool selector from the results, then use describe and call.
If authentication or connectivity prevents pi-mcporter from loading a server’s tools, the server still appears as a known MCP server. The result identifies the unavailable metadata and suggests authentication or connectivity recovery instead of reporting that the server doesn’t exist.
describe: get the full schema for a tool
Use when you know the selector but need to see its required parameters before calling.
{ "action": "describe", "selector": "linear.create_issue" }
安装
npx mcporter list # verify your servers are visible配置
{
"mcpServers": {
"linear": {
"baseUrl": "https://mcp.linear.app/mcp",
"headers": {
"Authorization": "Bearer ${LINEAR_API_KEY}"
}
},
"everything": {
"command": "npx -y @modelcontextprotocol/server-everything"
}
}
}