Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add liyecom/liye-ai --skill mcp-buildergit clone --depth 1 https://github.com/liyecom/liye-aiWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/liyecom/liye-ai/mcp-builder)<a href="https://agentmods.dev/skills/liyecom/liye-ai/mcp-builder"><img src="https://agentmods.dev/badge/skills/liyecom/liye-ai/mcp-builder/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/liyecom/liye-ai/mcp-builder"><img src="https://agentmods.dev/badge/skills/liyecom/liye-ai/mcp-builder.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00016 | $0.01031 |
| Opus 5 | $0.00008 | $0.00515 |
| Sonnet 5 | $0.00003 | $0.00206 |
| Haiku 4.5 | $0.00002 | $0.00103 |
Grade A, and why
mcp-builder scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
MCP Builder
来源: ComposioHQ/awesome-claude-skills 适配: LiYe OS 三层架构
指导创建高质量的 Model Context Protocol (MCP) 服务器,用于将外部 API 和服务集成到 LLM 中。
When to Use This Skill
当 Claude 需要帮助构建 MCP 服务器时:
- 设计 MCP 服务器架构
- 实现 Python 或 TypeScript MCP 服务器
- 定义工具接口和参数
- 处理错误和边界情况
- 测试和调试 MCP 集成
Core Capabilities
1. 架构设计
- MCP 协议理解
- 服务器结构规划
- 工具设计模式
- 资源管理策略
2. Python 实现 (FastMCP)
from fastmcp import FastMCP
mcp = FastMCP("my-server")
@mcp.tool()
def my_tool(param: str) -> str:
"""工具描述"""
return f"结果: {param}"
3. TypeScript 实现 (MCP SDK)
import { McpServer } from '@modelcontextprotocol/sdk/server';
const server = new McpServer({
name: 'my-server',
version: '1.0.0'
});
server.tool('my_tool', 'Tool description', {
param: { type: 'string', description: 'Parameter' }
}, async (args) => {
return { result: `Result: ${args.param}` };
});
4. 工具定义最佳实践
- 清晰的工具命名
- 详细的参数描述
- 类型安全的输入输出
- 错误处理模式
5. 测试与调试
- 单元测试策略
- 集成测试方法
- 日志和监控
- 性能优化
Usage Examples
示例 1: 创建数据库查询 MCP
用户: 帮我创建一个查询 PostgreSQL 的 MCP 服务器
Claude: [使用 mcp-builder 技能设计架构,生成代码框架]
示例 2: 集成第三方 API
用户: 我想让 Claude 能调用天气 API
Claude: [使用 mcp-builder 技能创建 weather-mcp-server]
示例 3: 扩展现有 MCP
用户: 给 SellerSprite MCP 添加新的工具
Claude: [使用 mcp-builder 技能分析现有结构,设计新工具]
Dependencies
- Python: fastmcp, mcp
- TypeScript: @modelcontextprotocol/sdk
LiYe OS Integration
业务域引用
此技能被以下业务域引用:
- 06_Technical_Development: MCP 生态扩展(主域)
已有 MCP 服务器参考
LiYe OS 已有 3 个 MCP 服务器可作为参考:
src/runtime/mcp/servers/amazon/sellersprite_server.pysrc/runtime/mcp/servers/knowledge/qdrant_server.pysrc/runtime/mcp/servers/data/(数据层)
三层架构位置
- 物理层 (本文件): Skills/00_Core_Utilities/development-tools/mcp-builder/
- 逻辑层索引: Skills/{domain}/index.yaml
- L3 指令层: .claude/skills/{domain}/mcp-builder/
Created: 2025-12-28 | Adapted for LiYe OS
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 143 lines · 16 tokens per session scan A e5a4bad3de5c
mcp-builder is a skill published in the GitHub repository liyecom/liye-ai (33 stars, last pushed 2d ago), licensed Apache-2.0. It adds 16 tokens to every session and 1,031 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
ensemble-beads-build
Drive an existing bead hierarchy to completion through the full builder, code-review, and close pipeline (Codex skill for /ensemble:beads-build).
dia-guide
Guides users through the V-Model workflow: takes stock of the project state, recommends the next phase skill, audits handoff state, and runs the Closing Handoff after a green security audit. Explicit user command for "where do I start", "what comes next", "wo bin ich gerade", "was kommt jetzt". Never for individual…
dia-realign
One entry point for repos that predate current DIA conventions: brownfield onboarding (existing codebase, legacy code, reverse engineer, we already have code) and legacy DIA upgrades (migrate v1/v2, upgrade DIA, FEATURE to FEAT, restructure backlog). Detects repo state, then runs a reverse walk, a script pass, or a…
business-analysis
Condenses a structured dialog about problem, users, and scope into a short BA record that feeds requirements engineering. Use when the user mentions "Business Analysis", "BA", "Problem Analysis", "Personas", "Define Scope", "Explore", "How might we", "Value Proposition", or starts a new project without a clear…
coding
Handoff and bug-capture skill: loads plan-context and design artifacts, reviews them critically against the real codebase, and keeps artifacts in sync during and after implementation. TDD is the default. Use for "implement", "code", "build feature", "realize plan-context", "Bug gefunden", "es gibt einen Fehler", "Fix…
security-audit
Formal full-codebase security audit producing a written report with prioritized findings: SAST, OWASP (incl. LLM Top 10), SCA, supply chain, Zero Trust. Trigger ONLY on explicit requests like "security audit", "OWASP audit", "Sicherheitsaudit", "dependency audit". NOT for PR-level checks, one-off questions, or generic…