README

OpenWebViewAI

OpenWebViewAI is a specialized, lightweight container designed to orchestrate Local AI interfaces. Built on native Delphi VCL and Microsoft Edge WebView2, it bridges the gap between the desktop operating system and local web-based AI tools. Unlike standard browsers designed for security isolation, OpenWebViewAI functions as a programmable workspace. It enables deep integration with trusted local servers (e.g., Generic AI Models, LLM Web UIs, Stable Diffusion interfaces), offering automation and asset management capabilities that streamline the AI workflow.

🎯 Core Purpose The primary goal is to provide a unified, high-efficiency environment for running offline AI models. By removing the friction between the OS and the local browser, users can automate repetitive tasks, manage prompts, and handle heavy media assets seamlessly across various local AI platforms.

⚠️ Usage Policy: This tool is strictly intended for productivity enhancement on user-authorized local systems. It is designed to facilitate local development and offline AI generation. It must not be used to circumvent security policies, scrape unauthorized data, or violate the Terms of Service of any third-party platforms.

🚀 Key Features

🖥️ Native Integration Portable & Fast: Zero-dependency deployment using memory-loaded libraries. No installation required. Resource Efficient: Native process management to keep the footprint low during heavy inference tasks. Secure Bridge: Token-based IPC (Inter-Process Communication) ensures safe data transfer between the host and the web interface.

🤖 Local AI Orchestration (Generic) Universal Connector: Designed to hook into various local AI backends (not limited to specific UIs), automating inputs and capturing outputs via WebSocket listeners. Smart Asset Handling: Intelligently processes images, videos, and text prompts, handling complex MIME types for seamless model ingestion. Workflow Automation: Automates the queueing and retrieval process, allowing for "batch" operations across different local models.

📊 Structural Visualization (Z-Graph) DOM Analysis: Instantly visualizes the structure of complex web interfaces as an interactive force-directed graph. Semantic Focus: Filters out layout noise to highlight functional elements (Inputs, Media, Controls), aiding in rapid UI comprehension and debugging.

🛠️ The "Taskbar" Runtime Embedded Scripting: A floating editor allows users to inject and execute custom automation scripts directly within the AI interface context. Scope Isolation: Ensures custom scripts run safely without polluting the global environment, resetting cleanly between executions.

📜 License Licensed under the Apache License 2.0. See the LICENSE file for full details. Disclaimer: This software is a local workflow utility provided "as is". The authors are not responsible for misuse or violations of third-party terms.