skills.rest
234194027-cpu
234194027-cpuCOMMUNITY
Explore all skills by this author →
GitHub

mcp-builder

Guide MCP server creation with Python FastMCP or Node TypeScript SDK.

Updated May 4, 2026
Search Tags:#typescript#evaluation#mcp#schema validation#tool design#fastmcp#llm agents
One-click install
npx skills add https://github.com/234194027-cpu/xianclaw --skill mcp-builder-234194027-cpu
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: mcp-builder
Source: https://github.com/234194027-cpu/xianclaw/tree/main/resources/openclaw/config/skills/mcp-builder
Command: npx skills add https://github.com/234194027-cpu/xianclaw --skill mcp-builder-234194027-cpu

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill solves the challenge of designing MCP (Model Context Protocol) servers whose tools are clear, discoverable, reliable, and effective for real LLM agent workflows.

Core Features & Use Cases

  • Workflow-first MCP design: Balance API endpoint coverage with higher-level workflow tools so agents can complete tasks efficiently.
  • Tool discoverability & schema clarity: Use strong naming conventions, concise tool descriptions, actionable error messages, and input/output schemas to improve tool selection and execution.
  • End-to-end quality and evaluation guidance: Cover setup, implementation, testing with MCP Inspector, and creating stable evaluation suites that verify whether agents can succeed using only the MCP tools.

Quick Start

Ask an AI to help you plan and implement an MCP server by following the phases and tool design guidance in this guide, then generate an evaluation XML with 10 complex read-only questions to validate tool usability.

Frequently Asked Questions about mcp-builder

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build an MCP server for LLM agents to use external APIs?▼

To build an MCP server for LLM agents, you need to design clear tool schemas, choose appropriate transports, and implement structured input/output with actionable error handling using Python (FastMCP) or Node/TypeScript (MCP SDK).

What is the best way to design MCP tool schemas for agent discoverability?▼

The best way to design MCP tool schemas for discoverability is to use strong naming conventions, concise tool descriptions, and structured input/output schemas to improve tool selection and execution by LLM agents.

How do I validate that my MCP server tools work correctly for agents?▼

You validate MCP server tools by running the MCP Inspector and executing scripted evaluations using read-only XML QA pairs to verify whether agents can successfully complete real-world tasks using only the provided tools.

Can I use FastMCP and TypeScript MCP SDK to create model context protocol servers?▼

Yes, you can use FastMCP for Python or the MCP SDK for Node/TypeScript to create model context protocol servers, balancing API endpoint coverage with higher-level workflow tools for efficient agent task completion.

How should I handle pagination and errors when building MCP tool servers?▼

When building MCP tool servers, handle pagination by defining its behavior in the tool design, and manage errors by implementing actionable error handling with proper annotations to guide agents during execution failures.

Related Skills

Explore More Software Engineering Skills→
montenokimontenoki

mcp-builder

Guide the creation of MCP servers with Python and Node/TypeScript SDKs.

Community
Advanced
MacPhobosMacPhobos

universal-main-mcp-builder

Create XML-formatted QA evaluations for MCP server tool interactions.

Community
Advanced
pacphipacphi
14

mcp-builder

Create MCP servers for LLM interaction with external services.

Community
Advanced
rahul-s-bhattrahul-s-bhatt

mcp-builder

Build MCP servers that connect AI agents to external services.

Community
Advanced