Project: SOCIAL PEAK - Full-Stack Web Development | Pool Deza
Details, tech stack, key features, and impact of SOCIAL PEAK
Modular educational web automation tool built on SeleniumBase and Flask, designed for academic and detection research purposes.
Social Peak is a high-performance modular web automation educational tool built in Python and powered by SeleniumBase and Flask, designed strictly for academic purposes and security research on anti-bot detection.
The main challenge of the project lay in bypassing advanced anti-bot detection mechanisms (such as Cloudflare, WebGL/Canvas fingerprinting, and IP blocks). To address this, I implemented a robust behavioral humanization module featuring pseudo-random mouse movements, realistic action delays, and a real-time rotating proxy validator and manager.
The system includes an interactive visual game-style launcher featuring an animated loading progress bar (0-100%), a minimizable System Tray integration, comprehensive i18n support powered by Flask-Babel, and the ability to run up to 75 parallel accounts simultaneously without session interference.
- RoleCreator & Lead Developer
- Year2025-2026
- Platforms6 Social Networks
- Parallelism75 Active Accounts
- EvasionStealth & Proxies
WebGL, Canvas fingerprinting, and user-agent evasion modules. Bulletproof stealth browser automation.
Multi-threaded and async architecture in Python to isolate and execute tasks on dozens of profiles simultaneously.
Rotates and tests proxies in real time with support for authenticated IPs (user:pass@host:port) to preserve account reputation.
Decoupled python modules for Instagram, Facebook, X/Twitter, TikTok, YouTube, and Kick, permitting autonomous updates.
Polished system-tray executable featuring customizable splash screen loader (0-100%) and auto-boot options.
Emits real-time task progress and browser telemetry through a dedicated WebSocket connection directly to the dashboard.
Injected customized client-side scripts inside SeleniumBase to override browser hardware readouts, reducing dynamic bot block rates by 95%.
Developed a centralized thread-safe profile isolator with distinct directory pools, ensuring zero session crossovers or cookies overlaps.
Optimized memory by systematically killing orphaned webdrivers, and packaged the runtime via PyInstaller into an optimized ~190MB bundle.
High-fidelity modular automation tool compiled for Windows (~190MB). Integrates SeleniumBase 4.0+ with customized stealth configurations, Flask for independent social network blueprints (Facebook, Instagram, Twitter/X, TikTok, YouTube, Kick), and pystray/customtkinter for the desktop launcher GUI. Features dynamic proxy validation in proxy.py and bidirectional WebSocket logging to the Universe controller.