CtrlK
BlogDocsLog inGet started
Tessl Logo

mcp-builder

Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP), Node/TypeScript (MCP SDK), or C#/.NET (Microsoft MCP SDK).

83

2.50x
Quality

73%

Does it follow best practices?

Impact

100%

2.50x

Average score across 3 eval scenarios

SecuritybySnyk

Advisory

Suggest reviewing before use

Optimize this skill with Tessl

npx tessl skill review --optimize ./.github/skills/mcp-builder/SKILL.md
SKILL.md
Quality
Evals
Security

Evaluation results

100%

72%

Build a TypeScript MCP Server for a Task Management API

TypeScript MCP server structure and modern SDK usage

Criteria
Without context
With context

Server naming format

0%

100%

registerTool API usage

0%

100%

Zod strict schemas

0%

100%

Tool annotations present

0%

100%

structuredContent in responses

0%

100%

CHARACTER_LIMIT constant

0%

100%

Truncation message

0%

100%

response_format parameter

50%

100%

Strict TypeScript config

100%

100%

No any type

100%

100%

Build succeeds

100%

100%

Project structure

50%

100%

100%

73%

Build a Python MCP Server for a Product Catalog API

Python MCP server with FastMCP and Pydantic v2

Criteria
Without context
With context

FastMCP import and init

0%

100%

Python server naming

0%

100%

Tool naming with service prefix

0%

100%

Pydantic v2 model_config

0%

100%

Pydantic v2 field_validator

0%

100%

Pagination metadata

40%

100%

response_format parameter

100%

100%

Shared error handler

0%

100%

Actionable error messages

62%

100%

No stdout logging

100%

100%

Async httpx usage

33%

100%

extra=forbid in models

0%

100%

100%

33%

Build a Bookstore MCP Server and Write Its Evaluations

Tool naming, annotations, and MCP evaluation creation

Criteria
Without context
With context

Tool names have service prefix

0%

100%

Tool names snake_case

100%

100%

All four annotations present

0%

100%

Tool descriptions include examples

75%

100%

Tool descriptions include return schema

100%

100%

Transport selection rationale

100%

100%

No SSE transport

100%

100%

Evaluation XML format

50%

100%

10 QA pairs

0%

100%

Questions are read-only

100%

100%

Answers are single verifiable values

100%

100%

Questions require multiple steps

100%

100%

Repository
microsoft/agent-skills
Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.