ck:use-mcp

Automate MCP tool discovery and execution in Gemini workflows with stdin piping.

1|Updated Mar 5, 2026
One-click install
npx skills add https://github.com/hotriluan/alkana_web --skill ck-use-mcp-hotriluan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ck:use-mcp
Source: https://github.com/hotriluan/alkana_web/tree/main/.opencode/skills/use-mcp
Command: npx skills add https://github.com/hotriluan/alkana_web --skill ck-use-mcp-hotriluan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates MCP-based tool discovery and execution within Gemini workflows, reducing manual orchestration and ensuring consistent context handling.

Core Features & Use Cases

  • Intelligent task orchestration: coordinates Gemini CLI and MCP subagents to locate and run appropriate MCP tools.
  • Safe execution with approvals: uses auto-approval flags and stdin piping to preserve MCP initialization and guard against unsafe scripts.
  • Fallback and recovery: gracefully falls back to the mcp-manager subagent when Gemini is unavailable, and suggests repair via ck:mcp-builder if needed.
  • No script generation required: enforces that no new scripts are created; relies on existing tools.

Use cases include applying MCP tools to perform server operations, manage configurations, and execute commands while preserving context budget and ensuring JSON-structured outputs.

Quick Start

Run an MCP operation by piping input through Gemini CLI with auto-approval, and fallback to the mcp-manager subagent when needed.

Frequently Asked Questions about ck:use-mcp

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

FAQPage Schema
How do I automate MCP tool discovery and execution within Gemini workflows?

Automating MCP tool discovery involves coordinating the Gemini CLI with MCP subagents to locate and run appropriate tools. This process uses stdin piping and auto-approval flags to ensure safe, deterministic execution while preserving your context budget.

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

When Gemini is unavailable, MCP execution gracefully falls back to the mcp-manager subagent. If further issues arise, it suggests using ck:mcp-builder to guide necessary repairs and restore normal tool orchestration.

Does MCP orchestration generate new scripts for tool execution?

No, MCP orchestration enforces a strict no-new-scripts constraint. It relies entirely on existing tools and deterministic execution via stdin piping to prevent unsafe script generation and maintain a secure environment.

How do I ensure JSON-formatted output when running MCP server operations?

To ensure JSON-formatted output during MCP server operations, pipe input through the Gemini CLI with auto-approval flags. This method enforces deterministic execution and structured outputs without requiring manual orchestration.

When should I use a fallback subagent for MCP management tasks?

You should use a fallback subagent for MCP management tasks when the primary Gemini environment becomes unavailable. This ensures continuous tool execution and prevents workflow disruption during server operations or configuration management.