Does this skill support both Python and TypeScript?

Yes, it provides specific implementation guides and best practices for both the Python FastMCP and the Node/TypeScript MCP SDKs.

What is 'agent-centric design' in this context?

It refers to designing tools specifically for AI usage, which includes building for complete workflows rather than simple API wrappers and providing actionable error messages that help the AI correct its own behavior.

How does this skill help with LLM context limits?

It teaches agent-centric design principles, such as returning high-signal information and providing configurable levels of detail to maximize the agent's context budget.

Can I use this to test my MCP server?

Yes, the skill includes a dedicated phase for creating comprehensive evaluation scenarios and quality checklists to ensure your server performs reliably under real-world tasks.

What is the Model Context Protocol (MCP)?

MCP is an open standard that allows Large Language Models (LLMs) to interact with external data and services through structured tools and resources.

MCP Server Builder

Name: MCP Server Builder
Author: ttmouse

byttmouse

•

API開発

Builds high-quality Model Context Protocol (MCP) servers for seamless integration between LLMs and external APIs.

This skill provides a comprehensive, four-phase framework for developing robust MCP servers using Python or TypeScript. It guides users through agent-centric design principles, implementation of workflow-optimized tools, and rigorous evaluation methodologies. By focusing on high-signal data transfer, actionable error messages, and context window efficiency, it ensures that AI agents can effectively interact with external services while maintaining operational safety and reliability.

主な機能

01Comprehensive error handling and natural language feedback systems

02Evaluation-driven development guides for testing server effectiveness

03Agent-centric design patterns focusing on workflows over raw API endpoints

04Step-by-step implementation workflows for Python (FastMCP) and TypeScript (MCP SDK)

05Tool optimization strategies for managing limited LLM context windows

0625 GitHub stars

ユースケース

01Building custom connectors for third-party SaaS platforms like Slack, GitHub, or Jira

02Standardizing the development of tool-calling capabilities across developer teams

03Integrating internal company APIs with Claude for automated backend workflows

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add ttmouse/skills mcp-builder

For use in Claude.ai and ChatGPT

Download Skill