guanyang/open-agent-hub

What it solves

It provides a centralized way to manage and activate modular capabilities—specifically skills, expert agent roles, and slash commands—for various AI coding assistants like Claude Code, Cursor, and Trae. Instead of manually copying prompts, users can dynamically link these capabilities to their project workspaces or global system configurations.

How it works

The project uses a zero-dependency CLI tool (oah) that manages standardized Markdown prompts with YAML frontmatter. It creates symlinks between the hub's library of components and the configuration directories used by supported AI assistants (e.g., .claude/, .cursor/). It also supports syncing these skills from upstream open-source community sources.

Who it’s for

Developers using AI coding assistants who want to extend their tools with a library of pre-defined expert workflows, modular prompting techniques, and custom commands without manual configuration.

Highlights

• Broad Compatibility: Supports a wide range of tools including Claude Code, Cursor, Trae, Gemini CLI, and others.
• Modular Library: Includes over 83 modular skills and dedicated specifications for Orchestrator, Evaluator, and Optimizer agent roles.
• Dynamic Management: Allows for project-level or global system-level activation of capabilities via a single command.
• Sync Capability: Built-in functionality to keep local skills updated from upstream sources.