use-mcp

Execute MCP operations via Gemini CLI with fallback to mcp-manager.

Updated Feb 27, 2026
One-click install
npx skills add https://github.com/haidonglethqb/CloudSchool --skill use-mcp
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: use-mcp
Source: https://github.com/haidonglethqb/CloudSchool/tree/main/.claude/skills/use-mcp
Command: npx skills add https://github.com/haidonglethqb/CloudSchool --skill use-mcp

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Coordinating MCP tool execution across environments can be brittle when relying on a single CLI and manual handoffs. This skill provides a robust orchestration that preserves the AI's context budget by executing MCP tasks through Gemini CLI with safe fallbacks to the mcp-manager subagent.

Core Features & Use Cases

  • Execution via Gemini CLI to preserve context budget.
  • Fallback orchestration to mcp-manager subagent when Gemini CLI is unavailable.
  • Guided fallback path via ck:mcp-builder when needed to repair tooling.
  • Use cases include automated MCP task execution in multi-environment setups with deterministic results and auditable trails.

Quick Start

Pass your task to the MCP tool using the Gemini CLI with stdin piping to initiate context-preserving execution.

Frequently Asked Questions about use-mcp

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

FAQPage Schema
How do I orchestrate MCP tools to preserve the AI context budget?

Orchestrate MCP tools by executing operations through Gemini CLI via stdin piping, which preserves the AI context budget by handling task execution externally and returning only deterministic results.

What happens to MCP task execution when Gemini CLI is unavailable?

When Gemini CLI is unavailable, MCP task execution automatically triggers fallback orchestration to the mcp-manager subagent to ensure operations continue safely without interrupting the workflow.

How does guided fallback repair broken MCP tooling environments?

Guided fallback repairs broken MCP tooling by routing execution to ck:mcp-builder, which provides structured troubleshooting steps to fix and restore unavailable MCP tools across environments.

Can I automate MCP workflows across multiple environments without writing new scripts?

Yes, you can automate MCP workflows across multiple environments without creating new scripts, because the orchestration enforces deterministic execution with safe fallbacks and avoids generating custom code.

Why does MCP tool execution fail during manual handoffs in multi-environment setups?

MCP tool execution fails during manual handoffs due to brittleness from relying on a single CLI, which this orchestration solves by coordinating Gemini CLI and subagent fallbacks automatically.