Awesome Omni Skills, developed by Diegosouzapw, is a free MCP server tool designed as an installable skill catalog and runtime for various AI coding assistants. Built with JavaScript/Node.js, it functions as an npm package, providing runtime surfaces for CLI, API, MCP, and A2A. Its core purpose is to connect to AI coding assistants like Claude Code, Cursor, and Gemini CLI, offering discovery, recommendation, and client-aware configuration tooling. This tool helps developers and users of AI coding assistants enhance their coding workflows.

What does Awesome Omni Skills connect to?

Awesome Omni Skills integrates with multiple AI coding assistants, including Claude Code, Cursor, OpenAI Codex, GitHub Copilot, Gemini CLI, and Snowflake Cortex Code. It provides essential runtime surfaces for CLI, API, MCP, and A2A operations. The MCP server component specifically assists with skill discovery, recommendation, and client-aware configuration tooling for these connected AI services. It also supports connections to Antigravity skills, broadening its utility across different AI programming environments.

Core features and tools

The MCP server component of Awesome Omni Skills exposes 87 distinct tools. These tools are categorized for operations such as routing, cache management, data compression, memory handling, skill execution, proxy services, and context source management. Specific tool definitions are found in files like tools.ts, memoryTools.ts, skillTools.ts, agentSkillTools.ts, gamificationTools.ts, pluginTools.ts, notionTools.ts, and obsidianTools.ts. The server supports stdio for IDEs, sse for browser clients, and streamable-http for multi-session HTTP.

Who is it for?

Awesome Omni Skills is intended for developers and users who interact with various AI coding assistants. It serves those who require a unified system to manage and execute AI coding skills across different platforms. The tool is particularly useful for individuals looking to enhance their productivity by integrating AI capabilities into their development workflows and maintaining shared project memory. Skip this if you require remote access to the MCP server without external authorization mechanisms already in place.

How to get started

No credentials are required for local loopback access, as the default authentication model is LOCAL_ONLY. For browser/agent clients needing an event stream, the sse transport protocol is used.

Use cases

• Exploring and inspecting semantic models, topics, views, and fields.
• Running queries against Omni's semantic layer.
• Creating and editing model definitions in YAML.
• Building a personal journaling practice.
• Enabling voice agent interactions.
• Syncing project or object context across AI coding tools for shared project memory.

Awesome Omni Skills provides a centralized catalog and runtime for AI coding assistant skills, supporting diverse clients and workflows. It helps developers unify AI tool interactions and manage project context effectively. Its architecture supports local operations by default, with external systems handling remote access policies.