mcp-integration-audit

Audit Claude MCP integration scoping and configuration for repositories.

14|1|Updated Jan 8, 2026
One-click install
npx skills add https://github.com/BlocUnited-LLC/mozaiks --skill mcp-integration-audit
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mcp-integration-audit
Source: https://github.com/BlocUnited-LLC/mozaiks/tree/main/.claude/skills/mcp-integration-audit
Command: npx skills add https://github.com/BlocUnited-LLC/mozaiks --skill mcp-integration-audit

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you avoid mis-scoped or unsafe MCP integrations by auditing how Claude MCP servers should be configured for a specific repo and team.

Core Features & Use Cases

  • Integration scoping guidance: Determines whether a given MCP integration should be configured as local, project, or user scope to match its purpose.
  • Secret-handling safety: Evaluates whether secrets can remain out of source control and identifies the safest approach for team workflows.
  • Repo planning support: Recommends whether a shared .mcp.json is appropriate and outlines a minimal safe configuration pattern.
  • Platform-specific concerns: Flags Windows stdio server pitfalls, especially around cmd /c npx usage, to reduce brittle setups.

Quick Start

Ask the AI: "Audit the MCP integration plan for the integration 'sentry' in this repository and recommend the smallest safe configuration pattern for the team."

Frequently Asked Questions about mcp-integration-audit

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

FAQPage Schema
How do I determine the correct scoping for a Claude MCP integration?

Claude MCP integration scoping depends on the integration's purpose: local for individual use, project for shared repository tooling, or user for cross-project access. The audit evaluates your specific integration to recommend the safest configuration scope.

Is it safe to commit a shared .mcp.json file to my repository?

A shared .mcp.json file is safe only if it excludes secrets and uses the smallest safe team config pattern. The audit evaluates whether your repository's planned MCP integrations can safely use a shared configuration without exposing sensitive credentials.

How do I manage secrets when configuring MCP servers for a team?

Secrets management for MCP servers requires keeping credentials out of source control. The audit evaluates your integration plan to identify the safest approach for team workflows, ensuring secrets remain protected while maintaining functional configurations.

Why does my npx-based MCP server fail on Windows with stdio?

Windows stdio server pitfalls often stem from npx execution issues, particularly around `cmd /c npx` usage. The audit flags these brittle setups and recommends configuration patterns that avoid common Windows-specific stdio problems.

When should I use project scope versus user scope for MCP integrations?

Project scope suits shared infrastructure tools like GitHub, Sentry, or database connections needed by the whole team, while user scope fits personal integrations. The audit determines the appropriate scoping by evaluating each integration's purpose and audience.

What is the smallest safe team configuration pattern for MCP integrations?

The smallest safe team configuration pattern uses a shared .mcp.json for non-sensitive project-scoped integrations while keeping secrets and user-specific credentials separate. The audit evaluates your integrations and outlines this minimal pattern for your team.