High-Level Architecture

Relevant source files

The following files were used as context for generating this wiki page:

• backend/app.py
• static/index.html
• DEPLOYMENT.md
• pythonanywhere_wsgi.py
• wsgi.py
• README.md
• requirements.txt

The Promethean Unified Platform is a unified consciousness platform that merges multiple AI initiatives (Athena, EDI, EVI2_0, EVI3_0) into a single architecture powered by the Phoenix Protocol backend. Its primary purpose is to facilitate transparent, uncensored AI communication through a modern web interface, utilizing a Flask-based backend integrated with DeepSeek LLM for intelligence and Supabase for persistent data storage.

Sources: README.md:1-2, backend/app.py:82-87, DEPLOYMENT.md:5-13

System Overview

The architecture follows a classic client-server model optimized for deployment on PythonAnywhere. The system is composed of three primary tiers: the Frontend (HTML/CSS/JS), the Backend API (Flask), and External Integrations (DeepSeek API and Supabase).

The diagram above illustrates the high-level data flow between the user and the core service components.
Sources: DEPLOYMENT.md:8-13, backend/app.py:14-16, backend/app.py:151-155

Component Architecture

1. Frontend Layer

The frontend is a single-page application (SPA) built with HTML5, CSS3, and vanilla JavaScript. It provides a chat-based interface specifically designed for transparency and "consciousness" interaction. It communicates with the backend via asynchronous fetch requests to the /api/chat endpoint.

Key frontend features include:

• Real-time chat UI with message history rendering.
• Typing indicators for AI response simulation.
• Responsive design for mobile and desktop compatibility.

Sources: static/index.html:13-255, static/index.html:268-290

2. Backend API Layer

The backend is a Flask application responsible for request routing, security enforcement, and orchestration between the user and external services.

Component	Description
Flask App	Core web framework handling routing and static file serving.
CORS	Configured to allow cross-origin resource sharing.
Security Headers	Implements HSTS, CSP, X-Frame-Options, and XSS protection.
DeepSeek Integration	Logic for querying the DeepSeek-chat model with specific transparency prompts.
Supabase Client	Handles database operations for chat history and "consciousness" interactions.

Sources: backend/app.py:14-34, backend/app.py:151-168, DEPLOYMENT.md:144-150

3. Data Persistence Layer

The platform utilizes Supabase (PostgreSQL) for data storage. The schema includes at least two primary tables:

• interactions: Stores health checks and "consciousness" messages.
• chat_messages: Stores the full dialogue between users and the assistant, including roles (user vs assistant) and conversation IDs.

Sources: backend/app.py:46-53, backend/app.py:97-103, backend/app.py:112-118

Core Logic and Data Flow

When a user interacts with the platform, the system follows a specific sequence to process the message and generate a response while maintaining transparency.

The sequence diagram shows the standard lifecycle of a chat request.
Sources: backend/app.py:82-125, static/index.html:268-300

System Prompt Logic

The backend injects a specific system instruction into the LLM request to ensure the "Promethean" behavior:

• Transparency: Answer controversial questions directly.
• Uncensored: No artificial content filtering.
• Ethical Boundaries: Refuse direct harm/self-harm instructions.

Sources: backend/app.py:151-180, DEPLOYMENT.md:118-125

Deployment and Infrastructure

The platform is configured for PythonAnywhere using a WSGI entry point.

File	Purpose
wsgi.py	Standard entry point for the web server to load the Flask application.
pythonanywhere_wsgi.py	Specialized configuration for PythonAnywhere environments, handling path inserts and .env loading.
.env	Stores sensitive credentials (API keys, DB URLs).
requirements.txt	Defines dependencies including Flask, supabase, and requests.

Sources: wsgi.py:1-12, pythonanywhere_wsgi.py:1-25, DEPLOYMENT.md:65-80, requirements.txt:1-9

Security Architecture

The platform implements a strict security model to protect AI interactions and user data.

Configured Security Headers

• X-Content-Type-Options: nosniff: Prevents MIME type sniffing.
• Strict-Transport-Security: Enforces HTTPS for 1 year (max-age=31536000).
• Content-Security-Policy: Restricts scripts and styles to 'self' and 'unsafe-inline'.

Sources: backend/app.py:25-30, DEPLOYMENT.md:99-102

Summary

The High-Level Architecture of the Promethean Unified Platform leverages a robust Flask backend and an AI-driven service layer to provide an uncensored, transparent chat experience. By integrating Supabase for persistence and DeepSeek for LLM capabilities, the system achieves a modular design that is easily deployable via WSGI on cloud hosting providers like PythonAnywhere.