What languages does the mcp-builder skill support?

It provides detailed implementation guides for both Python (using FastMCP) and Node/TypeScript (using the official MCP SDK).

Does it include help for data validation?

Yes, it covers implementing strict schema validation using Pydantic for Python and Zod for TypeScript to ensure robust data handling.

Can I use this for non-destructive read-only tools?

Yes, the skill includes specific guidance on using readOnlyHint and destructiveHint annotations to signal tool behavior clearly to the model.

How do I test my MCP server safely?

The guide recommends using a dedicated evaluation harness or running the server in a separate process like tmux to prevent the development environment from hanging.

How does this skill help optimize AI performance?

It emphasizes agent-centric design, teaching you how to consolidate operations into high-impact workflows and manage context budgets effectively.

MCP Server Builder

Name: MCP Server Builder
Author: EmaBilibili

byEmaBilibili

0•

API開発

Builds high-quality Model Context Protocol (MCP) servers to bridge LLMs with external APIs and services through optimized tool design.

The mcp-builder skill provides a comprehensive, step-by-step framework for developing robust MCP servers using either Python (FastMCP) or Node/TypeScript SDKs. It emphasizes agent-centric design principles, ensuring that tools are optimized for AI context windows, provide actionable error messages, and support complex workflows rather than just wrapping simple API endpoints. By following this guide, developers can create production-grade integrations that feature strict schema validation, efficient pagination, and rigorous evaluation-driven testing to ensure Claude can reliably interact with any external data source or service.

主な機能

01Structured input validation implementation using Pydantic and Zod schemas

02Agent-centric workflow design patterns to optimize LLM context usage

03Comprehensive error handling strategies with LLM-friendly natural language feedback

04Dual-language support for Python (FastMCP) and Node/TypeScript SDK implementations

050 GitHub stars

06Built-in evaluation framework for testing server reliability through complex QA pairs

ユースケース

01Developing high-performance MCP servers that handle large-scale data with smart truncation

02Building standardized toolsets for third-party SaaS integrations like CRM or project management platforms

03Creating custom API bridges to give Claude access to proprietary internal company tools

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add emabilibili/my-portfolio mcp-builder

For use in Claude.ai and ChatGPT

主な機能

01Structured input validation implementation using Pydantic and Zod schemas

02Agent-centric workflow design patterns to optimize LLM context usage

03Comprehensive error handling strategies with LLM-friendly natural language feedback

04Dual-language support for Python (FastMCP) and Node/TypeScript SDK implementations

050 GitHub stars

06Built-in evaluation framework for testing server reliability through complex QA pairs

ユースケース

01Developing high-performance MCP servers that handle large-scale data with smart truncation

02Building standardized toolsets for third-party SaaS integrations like CRM or project management platforms

03Creating custom API bridges to give Claude access to proprietary internal company tools