A cross-platform desktop GUI that unifies provider configuration, MCP server management, and skills synchronization across seven AI coding tools. It offers a local proxy for hot-switching and failover, plus usage cost tracking.
Project overview
Consolidates configuration management for multiple AI tools—including Claude Code, Codex, and Gemini CLI—into a single interface with a local proxy for format conversion and auto-failover.
Project type
AI Coding
Deployment
Refer to project documentation
License
MIT
Best for
Developers on Windows 10+, macOS 12+, or mainstream Linux distributions who need a desktop GUI to manage providers, proxy failover, MCP servers, and skills across multiple AI coding tools.
Key capabilities
Manage 50+ presets for 7 AI tools, import API keys with one click, switch instantly, and utilize system tray quick access.
Local proxy with hot-switching, format conversion, auto-failover, circuit breaker, and provider health monitoring.
Manage MCP servers, prompts (CLAUDE.md, etc.), and skills across supported apps with bidirectional sync and one-click install.
Track spending, requests, and tokens with trend charts, detailed request logs, and custom per-model pricing.
Browse, search, and restore conversation history across supported session sources and edit agent files.
Sync provider data across devices via Dropbox, OneDrive, iCloud, or WebDAV servers.
Limitations and risks
Linux setups using Wayland with NVIDIA drivers may experience unclickable web content areas and black-screen on resize, requiring an environment variable workaround.
Switching providers may require restarting the terminal or CLI tool for changes to take effect, with the exception of Claude Code.
Getting started
Setup is rated as easy. Pre-built installers are available for Windows, macOS, and Linux, alongside Homebrew and AUR packages. The first success path is to download and run the installer or use the package manager command, then launch the application.
Alternatives and comparisons
Provides an agentic skills framework enforcing test-driven development for coding agents, rather than a desktop application for provider and configuration management.
A system providing skills, rules, and security scanning across multiple AI agent harnesses, focusing on agent workflow enhancement rather than multi-tool configuration switching.
Composable engineering skills enforcing software fundamentals for AI coding agents, complementing agent workflows rather than centralizing provider and MCP management.