Zen functions as a Model Context Protocol (MCP) server that empowers CLI-based AI agents, such as Claude Code, Gemini CLI, or Codex CLI, to access and orchestrate a diverse team of AI models. It facilitates true AI collaboration with seamless conversation continuity, enabling complex workflows like multi-model code reviews, automated planning, and pre-commit validation to occur within a single, continuous conversation thread. By allowing models to 'remind' each other of past context and delegating subtasks based on their specific strengths, Zen ensures thorough analysis, extends context windows beyond individual model limits, and keeps the user in full control of their AI development team.
Key Features
01Multi-Model Orchestration and Automatic Model Selection
02Extended Context Windows by delegating to models with larger capacities
03Support for multiple AI providers including local models like Ollama
041 GitHub stars
05Comprehensive suite of core tools for code analysis, planning, and debugging
06Context Revival Magic for conversation continuity across AI context resets
