What are the requirements to run KognitiveKompanion?
Python 3.8+ (Python 3.12 recommended), PyQt5, and one of the backends: OpenAI API key, Ollama running locally, or AMD Ryzen AI compatible hardware. KDE Plasma 5 Desktop is optional.
@MagicUnicornInc
基本信息
KDE AI Interface and MCP Server Integration
工具列表暂未提供
当前收录暂时没有可展示的工具列表,不代表服务没有工具。请查阅项目文档了解支持的能力。
工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。
KognitiveKompanion is a modern, versatile AI interface for KDE and other desktop environments, designed to provide seamless interaction with AI backends including OpenAI, Ollama, and AMD Ryzen AI hardware acceleration. It is for users who want a desktop assistant with multi-backend support, screen capture, audio input, and conversation management.
Clone the repository, install dependencies with pip install -r requirements.txt, then run the application using one of the provided scripts: ./run_openai_gui.sh for OpenAI, ./run_kde_ai_interface.py for Ollama, or ./run_ryzen_ai_model.py for AMD Ryzen AI. Detailed backend configuration is documented in MULTI_BACKEND_GUIDE.md.
Python 3.8+ (Python 3.12 recommended), PyQt5, and one of the backends: OpenAI API key, Ollama running locally, or AMD Ryzen AI compatible hardware. KDE Plasma 5 Desktop is optional.
OpenAI API (GPT-4o, GPT-3.5-Turbo, etc.), Ollama for local models, and AMD Ryzen AI hardware acceleration.
Yes, the application works on other desktop environments as well; KDE Plasma is optional.
The application includes conversation management features for saving, loading, and exporting chat history.
The README does not specify data storage locations, but conversations can be saved and loaded locally via the UI.

A good sequence diagram can solve almost any communication problem in engineering. Volley lets your agent write them as plain text, with the real request and response payloads on every arrow. It checks its own syntax wi
Painless E2E Automation for Mobile and Web
MCP Toolbox for Databases is an open source MCP server for databases.

See and control a real Windows PC you own — from any MCP client, locally or remotely. Observe (UIA + screenshots), click/type/drag/scroll, launch apps, owner Live View. BYOH: your machine, your key.
🚀 The fast, Pythonic way to build MCP servers and clients.
评论