Do I need Gemini CLI to use this skill?

While Gemini CLI is the recommended primary execution method for speed and intelligence, the skill also provides direct TypeScript scripts and subagent fallbacks if Gemini is unavailable.

Can I manage multiple MCP servers at once?

Yes, this skill supports multi-server orchestration, allowing you to coordinate tools and resources across various configured servers from a single configuration file.

What is the Model Context Protocol (MCP)?

MCP is an open protocol that allows AI agents like Claude to securely connect to external tools, databases, and local resources through a standardized interface.

Where are the tool definitions stored?

The skill automatically saves discovered tools and their full schemas to assets/tools.json, enabling fast reference, offline browsing, and intelligent LLM analysis.

How does this skill help with context management?

It uses subagents and persistent tool catalogs to discover capabilities, ensuring that only the specific, necessary tool definitions are loaded into the main conversation window to save tokens.

MCP Server Manager

Name: MCP Server Manager
Author: Microck

byMicrock

•

開発者ツール

Manage, discover, and execute Model Context Protocol (MCP) capabilities across multiple servers while maintaining a clean context window.

概要

This skill provides a comprehensive toolkit for interacting with Model Context Protocol (MCP) servers, allowing Claude to efficiently discover and execute external tools, prompts, and resources. By utilizing intelligent tool selection and a persistent tool catalog, it enables multi-server management without overloading the conversation's context window. It is particularly useful for developers building complex AI agent workflows that require seamless integration with diverse external data sources and programmatic tool execution via Gemini CLI or subagent delegation.

主な機能

Intelligent LLM-driven tool selection based on task intent
Progressive discovery of MCP tools, prompts, and resources
Gemini CLI integration for high-performance tool execution
Persistent cataloging of capabilities to JSON for fast reference
Multi-server configuration and management from a single JSON file
81 GitHub stars

ユースケース

Reducing context token usage by delegating tool discovery to subagents
Orchestrating complex workflows across multiple MCP servers
Building and debugging custom MCP client implementations

概要

主な機能

Intelligent LLM-driven tool selection based on task intent
Progressive discovery of MCP tools, prompts, and resources
Gemini CLI integration for high-performance tool execution
Persistent cataloging of capabilities to JSON for fast reference
Multi-server configuration and management from a single JSON file
81 GitHub stars

ユースケース

Reducing context token usage by delegating tool discovery to subagents
Orchestrating complex workflows across multiple MCP servers
Building and debugging custom MCP client implementations