mcp-management

Discover and manage MCP servers to execute external tools via a centralized manager.

2|Updated Jan 9, 2026
One-click install
npx skills add https://github.com/NammDev/Goads-Krea --skill mcp-management-nammdev
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mcp-management
Source: https://github.com/NammDev/Goads-Krea/tree/main/.claude/skills/mcp-management
Command: npx skills add https://github.com/NammDev/Goads-Krea --skill mcp-management-nammdev

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @modelcontextprotocol/sdk, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

MCP servers expose capabilities that are costly to explore manually and can clutter the main agent context. This skill centralizes the discovery, analysis, and execution of MCP tools, prompts, and resources without polluting the primary workflow.

Core Features & Use Cases

  • Progressive disclosure of MCP capabilities: load only what's needed for a task.
  • Intelligent tool/prompt/resource analysis: identify the most relevant capabilities for a given objective.
  • Multi-server orchestration: manage and execute actions across configured MCP servers from a single config.
  • Context-efficient execution: delegate MCP operations to subagents to keep the main context lean.
  • Persistent catalog: automatically save discovered tools to assets/tools.json for fast reference and offline access.

Quick Start

Configure MCP servers in .claude/.mcp.json, install dependencies, and use the MCP Management CLI to discover capabilities and execute tools.

Frequently Asked Questions about mcp-management

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

FAQPage Schema
How do I manage multiple MCP servers without cluttering my main agent context?

Multi-server MCP management is handled by delegating tool discovery and execution to subagents, keeping the main context lean. You configure servers in a .claude/.mcp.json file and use a centralized manager to orchestrate actions without context pollution.

What is progressive disclosure of MCP capabilities and when do I need it?

Progressive disclosure of MCP capabilities means loading only the specific tools, prompts, or resources needed for a given task. You need it when manually exploring MCP servers becomes costly and starts cluttering your primary workflow context.

How to discover and execute MCP tools using a centralized CLI configuration?

Discover and execute MCP tools by defining servers in .claude/.mcp.json, then using the MCP Management CLI to list available capabilities. The skill analyzes tools, saves them to assets/tools.json, and safely calls them with structured lifecycles.

Does this MCP management approach work with the @modelcontextprotocol/sdk?

Yes, this MCP management approach works with the @modelcontextprotocol/sdk, which is the required dependency. It uses the MCP protocol to list tools, prompts, and resources, and to safely call tools across configured servers.

What's the best way to orchestrate tool execution across different MCP servers?

The best way to orchestrate tool execution across different MCP servers is using a centralized manager that reads from a single .mcp.json configuration. It performs intelligent tool analysis to identify relevant capabilities and delegates execution to subagents.

Can I save discovered MCP tools for offline reference and fast access?

Yes, you can save discovered MCP tools for offline reference. The skill automatically persists discovered tools, prompts, and resources into an assets/tools.json catalog, providing fast access without needing to repeatedly query the servers.