Does this skill support both Python and TypeScript?

Yes, this skill provides specific implementation guides, best practices, and project structure examples for both the Python (FastMCP) and Node/TypeScript SDKs.

What is an MCP server?

An MCP (Model Context Protocol) server is a standardized interface that allows Large Language Models like Claude to interact securely and efficiently with external data sources, APIs, and tools.

What are agent-centric design principles?

These principles focus on building tools optimized for how AI agents think, prioritizing high-signal information, workflow-based tools, and actionable error messages over simple API wrapping.

How can I verify the quality of my MCP implementation?

The skill includes language-specific quality checklists and a framework for creating 10 complex, realistic evaluation questions to test tool effectiveness.

How do I test my MCP server safely?

The guide recommends using an evaluation harness or running the server in a separate process like tmux to avoid hanging the main Claude Code process during stdio communication.

MCP Server Builder

Name: MCP Server Builder
Author: ai-wes

byai-wes

0•

API開発

Guides the development of high-quality Model Context Protocol (MCP) servers to bridge LLMs with external APIs and services.

This skill provides a comprehensive framework for building production-ready MCP servers using Python (FastMCP) or TypeScript. It emphasizes agent-centric design principles, such as optimizing for limited context and creating workflow-oriented tools rather than simple API wrappers. By following this guide, developers can implement robust infrastructure, handle complex errors gracefully, and create rigorous evaluation harnesses to ensure their MCP integrations perform effectively in real-world scenarios.

主な機能

01Support for Python (FastMCP) and Node/TypeScript SDK implementations

02Comprehensive guidelines for LLM-friendly error handling and response formatting

030 GitHub stars

04Agent-centric design patterns for context-efficient tool responses

05Framework for creating rigorous evaluation datasets to verify tool performance

06Structured workflow for tool selection, implementation, and refinement

ユースケース

01Creating specialized AI agents that require real-time access to external SaaS platforms

02Integrating internal company APIs with Claude for automated developer workflows

03Building custom database or filesystem connectors using the Model Context Protocol

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add ai-wes/glassbox-operator mcp-builder

For use in Claude.ai and ChatGPT

Download Skill