Evokore

关于

Evokore is a unified Model Context Protocol (MCP) server designed to supercharge AI assistants by centralizing a vast library of specialized knowledge. It aggregates, indexes, and dynamically serves over 200+ agent skills, prompts, and coding workflows, ensuring AI models like Claude, Cursor, Copilot, and Gemini have immediate access to contextually relevant information only when needed, keeping their context windows lean and operations efficient. This server significantly enhances the capabilities of various AI clients by providing a rich, dynamically loaded library of resources.

主要功能

• Progressive Disclosure: Keeps context windows lean by serving prompts only when requested by the LLM.
• 1 GitHub stars
• Massive Built-in Library: Includes 146 specialized developer plugins, the Hive Framework, the Manus 'planning-with-files' protocol, and custom orchestration workflows.
• Dynamic Skill Loading: Automatically scans and registers Markdown instructions as MCP Prompts, Resources, and Tools.

使用案例

• Providing context-aware responses and actions for AI agents across various platforms.
• Enhancing AI assistant capabilities with a vast library of specialized skills and prompts.
• Automating complex coding workflows for developers through AI-driven context.