mcp-management

Discover and execute MCP tools across configured servers with structured JSON output.

Updated Jan 24, 2026

Search Tags:#automation #mcp #execution #multi-server #discover #tool-management

Ready to run

npx skills add https://github.com/pikakit/agent-skills --skill mcp-management-pikakit

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: mcp-management
Source: https://github.com/pikakit/agent-skills/tree/main/.agent/skills/mcp-management
Command: npx skills add https://github.com/pikakit/agent-skills --skill mcp-management-pikakit

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires tsx, and includes scripts (resource) and assets (resource) components.

What problem does it solve?

Discover, analyze, and execute tools from configured MCP servers. Progressive disclosure: load only the tools you need, when you need them. Triggers on: MCP tools, server discovery, tool execution, multi-server.

Core Features & Use Cases

Progressive discovery with cached catalogs to assets/tools.json.
Deterministic, multi-method tool execution (Gemini CLI, Direct CLI, Subagent fallback).
Multi-server routing by server name with structured JSON responses.

Quick Start

Configure your MCP servers and run discovery to build the tools catalog.

Frequently Asked Questions about mcp-management

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

FAQPage Schema

How do I discover and list available tools across multiple MCP servers?▼

To discover MCP tools across multiple servers, you need a valid .mcp.json configuration file. The system automates server discovery, lists available tools, and caches the structured catalog to assets/tools.json for progressive loading.

What is the best way to execute MCP tools with automatic fallback?▼

Executing MCP tools is handled through a deterministic multi-method approach. It attempts execution via Gemini CLI, falls back to Direct CLI, and then uses a subagent fallback, returning structured JSON responses with status, data, and error fields.

Do I need Node.js to manage and route MCP server tools?▼

Yes, managing and routing MCP server tools requires Node.js version 18 or higher. You also need the tsx dependency and active access to your configured MCP servers to perform tool discovery and execution.

Can I route tool execution requests to a specific MCP server by name?▼

Yes, multi-server routing by server name is fully supported. You can target a specific MCP server from your configuration, and the system will execute the requested tool on that server and return the result in a structured JSON format.

How does progressive disclosure work when loading MCP tools?▼

Progressive disclosure for MCP tools means loading only the tools you need, when you need them. The system builds a cached catalog of available tools during discovery, preventing unnecessary overhead from loading all tools simultaneously.