skills.rest
luicabref97
luicabref97COMMUNITY
Explore all skills by this author →
GitHub

mcp:build-mcp

Create MCP servers with Python FastMCP and Node/TypeScript for external APIs.

Updated Apr 6, 2026
Search Tags:#typescript#python#tooling#mcp#api-integration#fastmcp#mcp-server
One-click install
npx skills add https://github.com/luicabref97/sushi-jungle-web --skill mcp-build-mcp-luicabref97
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: mcp:build-mcp
Source: https://github.com/luicabref97/sushi-jungle-web/tree/main/.agents/skills/mcp-build-mcp
Command: npx skills add https://github.com/luicabref97/sushi-jungle-web --skill mcp-build-mcp-luicabref97

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Guides engineers to design and implement MCP (Model Context Protocol) servers that expose well-structured, LLM-friendly tools so large language models can reliably perform real-world workflows across external APIs and services while minimizing context waste and ambiguous errors.

Core Features & Use Cases

  • Workflow-first tool design: Emphasizes building tools that represent complete tasks (e.g., schedule_event) rather than thin API wrappers.
  • Language-specific implementation guidance: Covers Python (FastMCP) and Node/TypeScript patterns including input validation (Pydantic/Zod), async I/O, and SDK integration.
  • Robust UX for agents: Defines concise vs detailed responses, actionable error messages, pagination and truncation strategies, and tool annotations for safe use by agents.
  • Evaluation and QA: Provides a process for creating realistic, verifiable evaluation suites and testing servers safely with an evaluation harness.

Quick Start

Create an MCP server that exposes concise, validated tools for my calendar API so an LLM can check availability and schedule events.

Frequently Asked Questions about mcp:build-mcp

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

FAQPage Schema
How do I build an MCP server that lets LLMs interact with external APIs?▼

To build an MCP server for external API integration, create workflow-oriented tools using Python with FastMCP or Node/TypeScript MCP SDKs. This involves defining input validation, handling async I/O, and exposing structured tools that enable LLMs to reliably perform real-world tasks across services.

What is the best way to design MCP tools for large language models?▼

Designing MCP tools for large language models requires a workflow-first approach, building tools that represent complete tasks like schedule_event rather than thin API wrappers. This minimizes context waste and ambiguous errors by providing concise or detailed outputs alongside actionable error messages.

Can I use FastMCP and TypeScript to create production-ready MCP servers?▼

FastMCP and TypeScript both support creating production-ready MCP servers. The implementation provides language-specific guidance covering input validation via Pydantic or Zod, async I/O handling, and SDK integration patterns to ensure reliable tool exposure for LLM agents.

How do I handle pagination and input validation when building MCP servers?▼

Handle pagination and input validation in MCP servers by applying language-specific SDK patterns using Pydantic or Zod. The implementation defines truncation strategies for pagination and enforces strict input validation to satisfy requirements for safe, agent-friendly interactions.

How do I test and evaluate an MCP server safely with an evaluation harness?▼

Test an MCP server safely by creating realistic, verifiable evaluation suites and running them through an evaluation harness. This QA process ensures tools operate safely with correct input validation and return actionable error messages before production deployment.

Why do my LLM agents fail when calling external API tools with ambiguous errors?▼

LLM agents fail with ambiguous errors when API tools lack actionable error messages and proper output formatting. Fix this by building MCP servers that define concise versus detailed responses, enforce input validation, and include tool annotations for safe agent use.

Related Skills

Explore More Software Engineering Skills→
felix-howardfelix-howard

ck:mcp-builder

Create MCP servers for LLM integration with FastMCP and Node/TypeScript SDKs.

Community
Advanced
leafcoderleafcoder
2

mcp-builder

Develop Model Context Protocol servers with testing frameworks for Python and TypeScript.

Community
Advanced
sandeshbagmaresandeshbagmare

async-python-patterns

Guide asyncio and concurrent programming patterns in Python.

Community
Advanced
Elevaria-biaElevaria-bia

mcp-builder

Design and implement MCP servers with Python FastMCP or Node/TypeScript MCP SDK.

Community
Advanced