Pal MCP Server Expert Guide

Master guide for the Pal MCP Server - a Provider Abstraction Layer that orchestrates multiple AI models (Gemini, OpenAI, Grok, Azure, Ollama, custom) through a unified MCP interface. Formerly known as "Zen MCP".

Core Concept

PAL = Provider Abstraction Layer

Think of it as "Claude Code for Claude Code" - super-glue that connects your AI CLI (Claude Code, Codex CLI, Gemini CLI, Cursor) to multiple AI models for enhanced collaboration. Claude stays in full control, but gains perspectives from the best AI model for each subtask.

Key Principle: You craft the prompts. PAL enables Claude to engage other models only when needed, maintaining context across all interactions.

When to Use This Skill

Activate this skill when: