HomeProjectsArticlesPK BoardAbout
Explore Projects

PAL MCP Server

Stale
GitHub Python NOASSERTION

Description

PAL MCP Server unifies Claude Code, GeminiCLI, and CodexCLI with multiple LLM providers (Gemini, OpenAI, OpenRouter, Azure, Grok, Ollama, and custom models) into a single collaborative MCP service.

Key Features

  • Multi-model orchestration: Claude Code coordinates with Gemini Pro, GPT-5, O3, and 50+ other models within a single prompt
  • CLI-to-CLI bridge (clink): spawn isolated subagent CLIs for code review, planning, and task delegation without polluting main context
  • Conversation threading across models: full context carries forward seamlessly between tools and models for complex multi-step workflows
  • Extended context windows: delegate massive codebases to Gemini (1M tokens) or O3 (200K tokens) beyond Claude's native limits
  • Context revival: even after Claude's context resets, other models can remind it of prior discussions without re-ingesting documents
  • Professional multi-pass code reviews with severity levels, confidence tracking, and consensus from multiple AI experts

Use Cases

💡 Multi-model collaborative code review with automated planning, implementation, and pre-commit validation
💡 Cross-model debate to reach deeper insights on architectural decisions and design trade-offs
💡 Context recovery across long-running sessions using external models as persistent memory bridges
💡 Role-specialized subagent delegation: spawn planner, code reviewer, or custom agents with specialized prompts

Strengths & Limitations

✅ Strengths

  • • High community interest (11.7k stars)

⚠️ Limitations

  • • No updates in over 8 months
  • • No clear open-source license

Tags

mcp multi-model claude-code llm agent

Categories

🔌 MCP Tools

Quick Start

Install PAL MCP via your package manager, configure it as an MCP server in Claude Code or your preferred CLI tool, then use natural language to orchestrate multiple models. Example: 'Perform a codereview using gemini pro and o3 and use planner to generate a detailed plan, implement the fixes and do a final precommit check.'
Visit GitHub

Related Projects

Awesome Claude Skills

74.3k · Python
Active A

A curated list of awesome Claude Skills, resources and tools for customizing Claude AI workflows.

mcptoolspython +1
  • · 1000+ production-ready skills - Covering 9 categories including document processing, dev tools, data analysis, marketing, and writing
  • · Cross-agent compatible - Works with Claude Code, Claude.ai, OpenAI Codex, Cursor, Gemini CLI, Windsurf and more
  • · Progressive loading architecture - Each skill loads only name and description (~100 tokens), full content on demand, hundreds coexist
View full details → View on GitHub

Open MCP Client

1.6k · TypeScript
Active A

An open-source MCP client that provides unified access to Model Context Protocol tools, enabling integration of any MCP server into AI applications with simplified tool calling.

mcptypescripttools +1
  • · Open-source MCP client for unified access to Model Context Protocol tools
  • · Integrates any MCP server into AI applications with simplified tool calling
  • · Separate frontend and agent architecture for clean separation of concerns
View on GitHub

Vibe-Trading

32.3k · Python
Active A+

Vibe-Trading is an open-source personal trading agent stack that combines LLMs, MCP tooling, and multi-agent workflows for algorithmic trading and backtesting.

ai-agentalgorithmic-tradingbacktesting +5
  • · One command empowers your trading agent with full quantitative trading capabilities via LLM + MCP toolchain
  • · Backtesting engine supporting 7 data sources (Tushare, OKX, CCXT, AkShare, etc.) with local caching
  • · Connector-first architecture supporting IBKR, Robinhood, Tiger, Alpaca, OKX, Binance paper trading
View full details → View on GitHub