Rulesync, developed by Dyoshikawa, is a command-line interface tool built with TypeScript and Node.js. It serves as a utility for managing configurations of various AI coding agents and generating files for MCP servers. Distributed as an npm package, via Homebrew, or as a single binary, Rulesync offers developers a centralized approach to AI skill management. It supports a diverse collection of AI clients, simplifying the setup and conversion of AI tool configurations across different platforms.

What does Rulesync connect to?

Rulesync connects to and manages configurations for a variety of AI development tools, known as AI coding agents. It functions as an AI configuration management CLI tool, specifically for AI Skill Management. Rulesync also supports the generation of configuration files for MCP servers, which these AI agents utilize. Confirmed compatible AI clients include Claude Code, Cursor, Copilot, Gemini CLI, Cline, Roo, Codex CLI, Windsurf, OpenCode, OpenClaw, Kilo, Junie, Factory Droid, Rovodev, Antigravity, JetBrains, Warp, Replit, Zed, Muse Code, and Devin.

Core features and tools

Rulesync exposes several callable tools for managing AI agent configurations. These include "rulesync init" for setup, "rulesync fetch owner/repo" to retrieve remote skills, and "rulesync generate --targets "*" --features "*"" for configuration output. Users can "rulesync import --targets <tool>" to bring in existing configurations or "rulesync convert --from <tool> --to <tool>,<tool>" to transform formats directly. Additional commands like "rulesync install", "rulesync generate --check", "rulesync generate --dry-run", "rulesync generate --watch", "rulesync docs [document]", and "rulesync docs --search <text>" are also available.

Who is it for?

Rulesync is designed for developers and engineers who need to manage configurations across multiple AI coding agents. It helps those who require a unified approach to handling AI rules, MCP configurations, and commands from a single source. This tool is valuable for individuals working with a diverse set of AI development environments and seeking to standardize their configuration processes. Skip this if you only use a single AI coding agent and do not require cross-tool configuration management.

How to get started

The server authenticates using an API key passed as an environment variable. Any necessary authentication for fetching skills relies on environment variables, with no information persisted.

Use cases

• Developers can manage AI rules, MCP configurations, and commands from a single set of unified rule files within a ".rulesync/" directory, then generate tool-specific configuration files for various AI coding agents.
• Users can import existing AI tool configuration files (e.g., "CLAUDE.md", ".cursorrules") into the unified ".rulesync/" format.
• Rulesync can directly convert configurations from one AI tool format to another without requiring the adoption of the ".rulesync/" source-of-truth workflow.

Rulesync provides a centralized solution for managing configurations across numerous AI coding agents and MCP servers. Its ability to unify, import, and convert configurations offers significant utility for developers. The tool supports a broad range of AI clients, making it a versatile option for complex AI development environments. Try importing an existing AI tool configuration file into the unified ".rulesync/" format.