mcp-builder

Guides building and evaluating MCP servers in TypeScript or Python.

6|1|Updated Sep 3, 2026
One-click install
npx skills add https://github.com/jasonzhu0922-sketch/agentloop --skill mcp-builder-jasonzhu0922-sketch
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mcp-builder
Source: https://github.com/jasonzhu0922-sketch/agentloop/tree/main/packages/agentloop-skills/skills/mcp-builder
Command: npx skills add https://github.com/jasonzhu0922-sketch/agentloop --skill mcp-builder-jasonzhu0922-sketch

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve? Building an MCP (Model Context Protocol) server that LLMs can actually use effectively requires careful tool design, schema validation, transport selection, and evaluation—this Skill provides a structured four-phase workflow covering all of it. ## Core Features & Use Cases - End-to-End MCP Development Workflow: Walks through research, implementation, review, and evaluation phases for building MCP servers in TypeScript (MCP SDK) or Python (FastMCP). - Language-Specific Reference Guides: Includes detailed implementation guides covering Zod/Pydantic schemas, tool registration, pagination, error handling, and transport configuration (stdio vs streamable HTTP). - Evaluation Harness: Ships Python scripts to create XML-based evaluation question sets and run automated LLM-based tests against your MCP server via stdio, SSE, or HTTP. - 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, and validate quality with 10 complex read-only evaluation questions. ## Quick Start Help me build an MCP server that integrates the GitHub API with tools for issues, pull requests, and repositories.

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 TypeScript?

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

How do I test whether my MCP server works well with LLMs?

Create an XML evaluation file with 10 complex, read-only questions and verifiable answers, then run the evaluation.py script against your server via stdio, SSE, or HTTP transport. The script uses Claude to attempt each question and reports accuracy, tool call counts, and per-task feedback.

Should I use stdio or streamable HTTP transport 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 multi-client scenarios; SSE is deprecated in favor of streamable HTTP.

What makes a good MCP tool name and description?

Use snake_case with a service prefix like github_create_issue to avoid conflicts when multiple servers are used together. Descriptions should precisely state functionality, document parameters and return schemas, and include annotations such as readOnlyHint and destructiveHint.

What are the requirements for MCP evaluation questions?

Questions must be independent, read-only, non-destructive, and complex enough to require multiple tool calls. Answers must be single verifiable values that remain stable over time, checked via direct string comparison.