mcp-tools

Select MCP plugins over built-in tools with fallback behavior.

15|Updated Apr 9, 2026
One-click install
npx skills add https://github.com/komluk/scaffolding --skill mcp-tools-komluk
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mcp-tools
Source: https://github.com/komluk/scaffolding/tree/main/skills/mcp-tools
Command: npx skills add https://github.com/komluk/scaffolding --skill mcp-tools-komluk

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enforces an MCP-first decision process to avoid ad-hoc tool selection, reduce failed plugin attempts, and ensure reliable fallbacks when plugins are unavailable or unauthenticated.

Core Features & Use Cases

  • Priority Decision Tree: Chooses MCP plugins first for supported operations and defines clear fallback behavior to built-in tools when plugins fail.
  • Plugin Quick Reference: Lists transport, key tool names, and agent mappings for common MCPs like context7, playwright, eslint, postgres-mcp, redis-mcp, github, and semantic-memory.
  • Operational Guidance: Documents auth, config, and usage patterns for semantic-memory and maps common operations (linting, DB queries, UI checks) to preferred MCPs and safe fallbacks.
  • Use Case: A developer requests a UI verification screenshot, the Skill routes to playwright if available, otherwise falls back to a manual browser check instruction.

Quick Start

Use the mcp-tools skill to prefer MCP plugins for operations and automatically fall back to built-in tools when plugins are unavailable or fail.

Frequently Asked Questions about mcp-tools

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

FAQPage Schema
How do I route operational tasks to MCP plugins instead of built-in tools?

To route operational tasks to MCP plugins, apply an MCP-first decision process that checks plugin availability and authentication, then automatically falls back to built-in tools when plugins are unavailable or fail.

What happens when an MCP plugin fails during database or UI operations?

When an MCP plugin fails during database or UI operations, the system applies defined fallback behavior to safely switch from the preferred plugin, such as playwright, to built-in tools or manual instructions.

Does the MCP-first approach work with semantic-memory and playwright plugins?

The MCP-first approach works with semantic-memory and playwright plugins by enforcing priority ordering, applying authentication checks, and verifying transport expectations before routing operational tasks.

How do I configure fallbacks for postgres-mcp and redis-mcp during database queries?

Configuring fallbacks for postgres-mcp and redis-mcp involves defining a priority decision tree that attempts the MCP plugin first for database queries, then safely routes to built-in tools if the plugin is unauthenticated or unavailable.

What is the best way to avoid ad-hoc tool selection when using MCP plugins?

The best way to avoid ad-hoc tool selection is enforcing an MCP-first decision process that maps operations like linting and UI checks to preferred MCPs, ensuring reliable fallbacks when plugins fail.

Why does my MCP plugin attempt fail for code inspection and linting operations?

MCP plugin attempts fail for code inspection and linting when authentication checks are not met, transport expectations mismatch, or the plugin is unavailable, triggering a fallback to built-in tools.