This repository provides a modular collection of tools, function pipes, and filters designed to enhance Open WebUI, transforming it into a powerful AI workstation for tasks ranging from academic research and image generation to music creation and autonomous agents. It targets users looking to extend their AI capabilities with specialized, integrated functionalities.

How It Works

The collection leverages Python scripts that integrate directly into the Open WebUI framework. Tools offer specific functionalities like arXiv or Pexels searches, image generation via Hugging Face or native Open WebUI backends, and music generation through ComfyUI. Function pipes enable complex workflows, such as the Planner Agent v2 for autonomous task execution with interactive error handling and the Flux Kontext ComfyUI Pipe for advanced image editing. Filters, like the Semantic Router and Prompt Enhancer, intelligently manage model selection and prompt quality.

Quick Start & Requirements

• Installation: Recommended via Open WebUI Hub (https://openwebui.com/u/haervwe) or manual copy-pasting of .py files into Open WebUI's Workspace.
• Prerequisites: Open WebUI version 0.6.0+ and Python 3.8+. Optional dependencies include ComfyUI, Mopidy, and various API keys (Hugging Face, Tavily, etc.).

Highlighted Details

• Planner Agent v2: An advanced autonomous agent with specialized models, interactive error handling, and real-time progress visualization via Mermaid diagrams.
• ComfyUI Integration: Seamless integration with ComfyUI workflows for advanced image and music generation/editing.
• Academic Focus: Tools like arXiv Search and arXiv Research MCTS Pipe facilitate academic research and paper discovery.
• Autonomous Agents: Supports integration with Letta agents and features the Planner Agent for complex, multi-step task execution.
• MCP Pipe: Integrates the Model Context Protocol for connecting AI systems with data sources and tools, supporting Python-based MCP servers.

Maintenance & Community

The project is part of the broader Open WebUI ecosystem. Support and bug reporting are handled via GitHub issues.

Licensing & Compatibility

The repository is licensed under the MIT License, permitting commercial use and integration with closed-source applications.

Limitations & Caveats

The MCP Pipe currently only supports Python-based MCP servers, with NPX or other server types requiring separate advanced pipeline configurations. Some tools require specific API keys and external service setups (e.g., Perplexica, ComfyUI).