概要
This skill provides a robust framework for integrating Model Context Protocol (MCP) servers into the Claude Code environment, enabling AI agents to connect to external tools and data sources seamlessly. It addresses the challenge of context bloat by utilizing a progressive disclosure model, where tool definitions and resources are only loaded when strictly necessary. With support for multi-server management, persistent tool cataloging, and intelligent LLM-driven tool selection, this skill allows developers to orchestrate complex workflows across various APIs and databases while keeping the main conversation context clean and performant.
