mcp-builder

Guides building and evaluating MCP servers in Python or TypeScript.

Updated Apr 17, 2026
One-click install
npx skills add https://github.com/Syedyasir001/RVULibPass --skill mcp-builder-syedyasir001
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: mcp-builder
Source: https://github.com/Syedyasir001/RVULibPass/tree/main/.agent/skills/library/mcp-builder
Command: npx skills add https://github.com/Syedyasir001/RVULibPass --skill mcp-builder-syedyasir001

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires anthropic, mcp, and includes scripts (resource) and references (resource) components.

What problem does it solve? Building MCP servers that LLMs can actually use effectively requires careful tool design, schema validation, transport selection, and testing—this Skill provides a structured four-phase workflow covering research, implementation, review, and evaluation. ## Core Features & Use Cases - Phased Development Workflow: Walks through research and planning, implementation, code review, and evaluation creation for MCP servers. - Language-Specific Guides: Includes reference documentation for Python (FastMCP with Pydantic) and TypeScript (MCP SDK with Zod) implementations, plus MCP best practices. - Evaluation Harness: Ships Python scripts to connect to MCP servers over stdio, SSE, or HTTP and run XML-based QA evaluations against them using Claude. - Use Case: You need to expose your company's internal API to LLM agents. Use this Skill to design well-named tools with proper annotations, implement the server in TypeScript, then generate 10 complex evaluation questions and run the harness to measure how well an LLM can use your server. ## Quick Start Help me build an MCP server in TypeScript that wraps the GitHub API with tools for issues and pull requests.

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 in Python?▼

Use the FastMCP framework from the official MCP Python SDK. Define Pydantic models for input validation, register tools with the @mcp.tool decorator including annotations like readOnlyHint, and run the server over stdio or streamable HTTP transport.

How do I build an MCP server in TypeScript?▼

Use the MCP TypeScript SDK's McpServer class and registerTool method with Zod schemas for input validation. Provide title, description, inputSchema, and annotations for each tool, then connect via StdioServerTransport for local use or StreamableHTTPServerTransport for remote deployment.

Should I use stdio or streamable HTTP for my MCP server?▼

Use stdio for local integrations and command-line tools where the server runs as a subprocess of a single client. Use streamable HTTP for remote servers, web services, and scenarios requiring multiple simultaneous clients.

How do I test whether an LLM can use my MCP server?▼

Create an XML evaluation file with 10 complex, read-only questions that have single verifiable answers, then run the evaluation.py script with your transport type. The harness connects to your server, runs each question through Claude, and reports accuracy, tool call counts, and per-task feedback.

What makes a good MCP tool description?▼

A good tool description concisely states functionality, documents each parameter with constraints and examples, and specifies the return value schema. Include usage examples showing when to use the tool and when to use alternatives, plus expected error messages.

Why do my MCP evaluation questions need stable answers?▼

Evaluation answers must be verifiable by direct string comparison, so they cannot rely on dynamic state like current issue counts or member numbers. Base questions on historical, closed data such as completed projects or archived repositories so answers never change over time.