mcp-integration

Manage Docker-backed MCP server lifecycle with transport auto-detection.

21|4|Updated Feb 19, 2026
One-click install
npx skills add https://github.com/tylerjrbuell/reactive-agents-ts --skill mcp-integration-tylerjrbuell
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mcp-integration
Source: https://github.com/tylerjrbuell/reactive-agents-ts/tree/main/.agents/skills/mcp-integration
Command: npx skills add https://github.com/tylerjrbuell/reactive-agents-ts --skill mcp-integration-tylerjrbuell

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provides clear, deterministic patterns for running and shutting down Docker-backed MCP servers so agents and tools do not leave orphaned containers, mis-detect transports, or collide when multiple agents run the same server.

Core Features & Use Cases

  • Transport Auto-Detection: Races stdio connect and stderr URL detection to automatically switch between stdio and HTTP transports.
  • Two-Phase Container Naming: Uses probe and managed container name phases to avoid naming collisions between concurrent agents.
  • Deterministic Cleanup: Enforces docker rm -f as the only reliable way to stop and remove containers and documents calling cleanupMcpTransport(serverName) from agent lifecycle hooks.
  • Use Case: Ideal for developers adding MCP client support to local tooling, Cortex server imports, or CI tests that mock Docker subprocesses.

Quick Start

Start the MCP server container with a standard MCPServerConfig, let the client auto-detect stdio or HTTP transport, and always call cleanupMcpTransport(serverName) when finished.

Frequently Asked Questions about mcp-integration

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

FAQPage Schema
How do I auto-detect stdio or HTTP transport for a Docker MCP server?

Transport auto-detection for Docker MCP servers races stdio connect and stderr URL detection to automatically switch between stdio and HTTP transports based on which endpoint responds first.

Why do my Docker MCP containers collide when multiple agents run concurrently?

Docker MCP container collisions occur because of naming conflicts. Using a two-phase probe and managed container naming strategy avoids naming collisions between concurrent agents running the same server.

What is the best way to stop and remove Docker MCP containers without leaving orphans?

The best way to stop Docker MCP containers without leaving orphans is enforcing docker rm -f, which provides deterministic container termination when paired with agent lifecycle hooks.

How do I clean up MCP server containers from agent lifecycle hooks?

Clean up MCP server containers by calling cleanupMcpTransport(serverName) from agent lifecycle hooks, ensuring deterministic container termination and removal using docker rm -f when the agent finishes.

Can I use this MCP transport management for CI testing scenarios with Docker subprocesses?

Yes, this MCP transport management applies to CI testing scenarios that mock Docker subprocesses, providing reliable container lifecycle control and transport detection for local or continuous integration environments.

Does this MCP client integration support Cortex server imports?

Yes, this MCP client integration supports Cortex MCP server configuration, allowing developers to add reliable Docker-backed server lifecycle management and transport auto-detection to Cortex imports.