Summary

OpenCluely: open-source desktop app delivering invisible, real-time AI assistance for technical interviews (DSA, OAs, CP). Targets candidates seeking an edge via process disguise, AI-powered question analysis, and customizable interfaces, ensuring complete stealth from interview monitoring tools.

How It Works

Stealth achieved via click-through transparent overlays and disguised processes, evading screen-sharing/recording detection. Core intelligence leverages Google Gemini for direct image analysis, bypassing OCR. Optional voice input via Azure Speech or local OpenAI Whisper, with context memory maintaining conversational continuity. UI features glassmorphism and draggable elements for customizable, unobtrusive experience.

Quick Start & Requirements

Installation streamlined via ./setup.sh script post-cloning. Key requirement: Google Gemini API key. Optional voice features necessitate Azure Speech credentials or local OpenAI Whisper setup (may require ffmpeg, sox). Node.js 18+ recommended. Demo video link provided: https://github.com/user-attachments/assets/896a7140-1e85-405d-bfbe-e05c9f3a816b.

Highlighted Details

• Stealth: Process disguise, click-through windows, invisibility to screen-sharing platforms (Zoom, Teams, Meet, Discord).
• AI Integration: Direct Gemini image analysis, multi-language support (C++, Python, Java, JS, C), session context memory.
• User Interface: Glassmorphism design, draggable/resizable windows, floating command bar, syntax highlighting.
• Functionality: Global shortcuts, optional voice commands, multi-monitor support.

Maintenance & Community

Active development indicated with daily improvements. Acknowledges community contributions and feedback. Developed by TechyCSR. Specific community channels or detailed contributor information absent from README.

Licensing & Compatibility

Licensed under permissive MIT License. Allows broad compatibility, including commercial use and closed-source integration without copyleft restrictions.

Limitations & Caveats

Active development phase; features like auto-hiding during screen share and multi-model support are in progress. Users must manage external API keys (Gemini, optional Azure) and ensure system dependencies (ffmpeg, sox) for voice features. macOS requires explicit screen recording permission grants. Ethical disclaimer emphasizes user responsibility for interview policy adherence.