SkillAgentSearch skills...

xingtu-mcps

生产级 MCP 服务集合:协议先行、可独立部署、跨 MCP 宿主复用的标准 server · Production-ready MCP servers, protocol-first, host-agnostic and independently deployable

Install / Use

claude mcp add xingtu1996 -- npx -y github:xingtu1996/xingtu-mcps

If the server publishes to npm under a different name, use that package instead — check the repo README.

About this skill
🔌

MCP Server

Model Context Protocol server

Quality Score

69/100

Category

Operations

Supported Platforms

Claude Code
Claude Desktop

Tags

Our assessment of xingtu-mcps

xingtu-mcps scores 69/100 on our quality scale, 699th of 757 Operations skills we index.

Its MCP Server is 2.3 KB long, well organised into 11 sections with 1 code example: moderately detailed.

It has 3 GitHub stars, so there is little community track record yet; judge it on its content.

Substance
20/30
Structure
17/20
Description
15/15
Adoption
3/20
Freshness
15/15

Maintenance, license and trust

  • The repository was last updated 3 days ago, so xingtu-mcps is actively maintained.
  • It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
  • Its trust signals score 92/100, with 1 caution from licensing, adoption, age or documentation. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.

xingtu-mcps compared with similar skills

All 4 of these similar skills score higher than xingtu-mcps; compare them before choosing.

SkillScoreStarsUpdatedFormat
xingtu-mcps (this skill)by xingtu19966933d agoMCP Server
Agent-Reachby Panniantong100108.9k1d agoCLAUDE.md
headroomby headroomlabs-ai10078.9ktodayCLAUDE.md
CowAgentby zhayujie10047.8k1d agoCLAUDE.md
Scraplingby D4Vinci10097.0ktodayMCP Server

Frequently asked questions

How do I install xingtu-mcps?
Run claude mcp add xingtu1996 -- npx -y github:xingtu1996/xingtu-mcps. The install tabs above show the steps for each supported agent.
Which AI agents does xingtu-mcps work with?
It is written for Claude Code and Claude Desktop, as a MCP Server file. Other agents that read the same format can often use it too.
Is xingtu-mcps safe to use?
It is MIT-licensed and scores 92/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
Is xingtu-mcps still maintained?
The repository was last updated 3 days ago, so xingtu-mcps is actively maintained.

xingtu-mcps · 生产级 MCP 服务集合

行途开源矩阵的 MCP 服务资产仓:收录可独立部署、可被任意 MCP 客户端复用的 MCP server。协议先行,不绑死某个 Agent 宿主——换客户端不用重写。

这是什么

Agent 只靠文本提示只能「说」,MCP(Model Context Protocol)让它「做」。本仓把散落在真实工程里的可复用工具能力收敛成标准 MCP server:任何支持 MCP 的宿主(Claude Code、Claude Desktop、Cursor、CodeBuddy 等)都能通过同一种方式发现和调用。

核心价值

  1. 协议先行:所有 server 走 MCP 规范(JSON-RPC 2.0 + 工具 schema),不绑死某个 Agent 宿主,换客户端不用重写。
  2. 标准化起步:每个 server 从 servers/_TEMPLATE 起步,统一目录与 README 结构(安装 / 配置 / 工具列表),把「起一个 server」的成本压到最低。
  3. 可独立部署:每个 server 独立运行,stdio 或 HTTP(S) 传输都可挂载,本地、远端皆可。
  4. 可审计:工具带参数 schema 声明,调用有明确的入参 / 出参边界,Agent 调用行为可记录、可回放。
  5. 与 harness 打通:作为子模块被 xingtu-harness 聚合,场景安装时按需拉取,而不是全量堆进 Agent 环境。

目录结构

xingtu-mcps/
├── servers/
│   └── feishu-bot/          # 飞书长连接机器人参考实现
├── SELF.md                  # 仓内专业自述(定位 / 边界 / 概念)
├── CONTRIBUTING.md
└── LICENSE

当前服务

feishu-bot · 飞书长连接机器人参考实现

servers/feishu-bot/ 是一个可运行的飞书 WebSocket 长连接机器人骨架:接收群聊消息、按命令回复、带 LLM 对话与上下文管理。所有密钥(飞书 AppID / Secret、LLM Key)只从环境变量读取,代码与配置不落盘。

| 文件 | 说明 | |---|---| | config.py | 配置(密钥全走环境变量,无硬编码) | | bot_longconn.py | 主入口:长连接 + 消息分发 | | commands.py | 快捷命令处理(/help /status /tasks 等) | | context.py | 群聊上下文保留与过期 | | llm.py | LLM 对话封装(OpenAI 兼容格式) | | debug_ws.py | 长连接连通性调试脚本 |

与相邻仓的边界

| 对比项 | xingtu-mcps | xingtu-tools / xingtu-cli | xingtu-skills | |---|---|---|---| | 能力形态 | 独立运行的 server 进程 | 本机脚本 / 命令行 | 文档式技能(SKILL.md + 辅助脚本) | | 调用方式 | Agent 经 MCP 协议调工具(带 schema) | 人或 Agent 直接执行命令 | Agent 按需读取技能说明后自行执行 | | 边界 | 工具即服务,有生命周期、可远端部署 | 无协议,进程内跑在宿主机器上 | 教 Agent「怎么想怎么做」,不产副作用工具 |

一句话:skills 给方法,tools/cli 给本机命令,mcps 给跨宿主的服务化工具。三者不重叠,可同时挂载。

新增 server

  1. cp -r servers/_TEMPLATE servers/<name>。
  2. 实现 tools / resources / prompts,写独立 README(安装 / 配置 / 工具列表)。
  3. 质量门禁:每个工具可跑通一次真实调用、参数 schema 校验通过;有副作用的工具必须有幂等或保护设计;README 未列出的工具不作为发布资产。

与聚合仓的关系

关于作者 · 行途 XingTu

License

MIT License。

Related Skills

View on GitHub
GitHub Stars3
CategoryOperations
Updated3d ago
Forks0

Languages

Python

Trust signals

92/100

From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.

1 low