What is a passing score for the MCP Directory?

A score of 90-100 (Grade A) is considered production-ready and highly eligible for MCP Directory submission, while scores below 75 indicate a need for significant improvement.

How does it measure token efficiency?

It analyzes tool schemas and response structures to identify unnecessary verbosity that could lead to higher latency or cost during LLM interactions.

Can this skill audit non-Python MCP servers?

While specifically optimized for the FastMCP Python framework, the logic and quality checklist are applicable to any MCP server implementation following the standard protocol.

Does it automatically fix the code issues it finds?

It provides a detailed report and actionable recommendations. You can use subsequent commands or skills like 'skill-creator' to implement the suggested fixes based on the report.

FastMCP Quality Reviewer

Name: FastMCP Quality Reviewer
Author: theflysurfer

bytheflysurfer

•

API Development

Audits and scores Model Context Protocol (MCP) servers to ensure production readiness and compliance with Anthropic directory standards.

About

The FastMCP Quality Reviewer is a specialized skill designed to automate the auditing process for MCP servers built using the FastMCP framework. It evaluates server code against official Anthropic MCP Directory policies, checking for critical factors such as tool description clarity, robust error handling, schema annotations, and token efficiency. By providing a comprehensive quality report with a numerical score (0-100) and categorized recommendations, it helps developers identify 'must-fix' issues and optimize their tools for better LLM performance and directory eligibility.

Key Features

Verification of test coverage and usage documentation
Validation of tool descriptions for length and clarity compliance
Token efficiency auditing to minimize response overhead
1 GitHub stars
Automated scoring based on Anthropic MCP Directory Policy
Detailed error handling and exception management analysis

Use Cases

Quality assurance during the development of new FastMCP tools
Improving token efficiency and performance of existing MCP implementations
Pre-publication audit before submitting servers to the official MCP directory

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add theflysurfer/claude-skills-marketplace julien-dev-tools-fastmcp-quality-review

For use in Claude.ai and ChatGPT

Download Skill

GitHub