Athena Integration

Relevant source files

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

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

The Athena Integration represents a core component of the Promethean Unified Platform, serving as a pillar within a unified consciousness architecture that merges multiple AI entities including EDI, EVI2_0, and EVI3_0. It operates within the Phoenix Protocol backend, designed for deployment on PythonAnywhere to provide a secure, transparent, and uncensored AI interaction environment.

Sources: README.md:1-2, DEPLOYMENT.md:5-13

System Architecture

The integration follows a tiered architecture comprising a Flask-based backend, a Supabase PostgreSQL database for persistent storage, and a DeepSeek LLM integration for cognitive processing. The system emphasizes "full transparency" and "uncensored" communication while maintaining strict ethical guardrails against direct harm.

Backend Infrastructure

The backend is built using Flask with standard WSGI entry points for production environments. It utilizes specific security headers to enforce HTTPS, prevent clickjacking, and mitigate XSS attacks.

The diagram illustrates the flow from the user interface through the secure Flask backend to the processing and storage layers.
Sources: backend/app.py:12-28, DEPLOYMENT.md:80-87, wsgi.py:4-13

AI Cognitive Logic and LLM Integration

The integration utilizes the DeepSeek API to provide uncensored responses. The system prompt is engineered to ensure the AI remains transparent, especially regarding controversial topics, while rejecting instructions for self-harm or violence.

Interaction Flow

When a message is received, the system performs a multi-step operation:

1. Validates the input message.
2. Logs the user's intent to the chat_messages table in Supabase.
3. Forwards the prompt to the DeepSeek API with a specific system instruction.
4. Records the AI's response in the database.
5. Returns the response to the user with a transparency metadata note.

The sequence shows the synchronous interaction between the user interface, the backend logic, and external services.
Sources: backend/app.py:80-128, DEPLOYMENT.md:90-95

Data Models and Storage

The platform uses Supabase for interaction logging and conversation history. Data is partitioned into two primary tables to track general consciousness interactions and structured chat sessions.

Table Schema Summary

Table	Field	Type	Description
interactions	user_id	String	Unique identifier for the user (default: 'anonymous')
interactions	message	Text	The raw input from the user
interactions	emotion_data	JSON	Metadata regarding the interaction status
chat_messages	conversation_id	String	Identifier to group related messages
chat_messages	role	String	'user' or 'assistant'
chat_messages	content	Text	The actual text content of the message

Sources: backend/app.py:44-55, backend/app.py:94-110

API Implementation

The Athena Integration exposes several RESTful endpoints for frontend interaction and health monitoring.

Core Endpoints

• POST /api/chat: The primary interface for uncensored AI communication. It requires a JSON body containing the message and an optional conversation_id.
  • Header: user-id (optional)
  • Response: Includes the AI response and a safety_note regarding the absence of harm instructions.
• POST /api/consciousness/interact: A specialized endpoint for processing "consciousness" messages, which are logged with specific emotion_data.
• GET /api/conversations: Retrieves the full message history for a specific user-id.

Sources: backend/app.py:38-40, backend/app.py:65-67, backend/app.py:131-133, DEPLOYMENT.md:65-78

Security and Deployment Configuration

Deployment is optimized for PythonAnywhere using environment variables to manage sensitive credentials like DEEPSEEK_API_KEY and SUPABASE_ANON_KEY.

Security Headers Configuration

The application explicitly sets headers to ensure a secure environment:

• Strict-Transport-Security: Enforces HTTPS for 1 year.
• X-Frame-Options: Set to DENY to prevent clickjacking.
• Content-Security-Policy: Restricts script and style sources to 'self' and 'unsafe-inline'.

Sources: backend/app.py:19-27, pythonanywhere_wsgi.py:16-21, DEPLOYMENT.md:43-58

Summary

The Athena Integration provides a robust framework for transparent AI communication within the Promethean Unified Platform. By combining the uncensored processing capabilities of DeepSeek with a secure Flask/Supabase architecture, it enables high-integrity interactions that are both logged for history and protected by industry-standard security protocols.