P.DE.I Framework

Personal Data-driven Exocortex Interface

The P.DE.I Framework is a sophisticated system designed to create a personalized AI extension of yourself. By analyzing your code repositories, commit history, and communication patterns, it creates an "Exocortex" that mirrors your coding style, values, and problem-solving methodologies.

🧠 Core Concepts

• Identity Mirroring: The AI adopts your persona, defined in personalities/, ensuring interactions feel like an extension of your own thought process.
• Forge Theory: An adaptive learning mechanism that tunes the AI's responses based on acceptance rates and retention bias.
• Secure Context: Built with security in mind, utilizing SSH/GPG for operations and local execution.

📂 Project Structure

• pdei_core/: Contains the core logic and the learning_manifest.json which directs the AI to your source code for training.
• personalities/:
  • users/: User-specific profiles (e.g., james_the_giblet.json).
  • instances/: AI personality instances (e.g., exocortex_universal.json).
• integrations/: Client interfaces.
  • vscode-pdei/: The Visual Studio Code extension to bring the Exocortex into your editor.
• scripts/: Utilities for initialization and model management.

🚀 Getting Started

1. Start the Daemon

The core of the system runs as a local daemon. Double-click start_daemon.bat or run:

start_daemon.bat

This will:

• Launch the P.DE.I Daemon.
• Open the Web Interface at http://localhost:8000.
• Initialize the Exocortex with the default personality.

2. VS Code Integration

To connect your editor to the Exocortex:

1. Open the integrations/vscode-pdei folder.
2. Install dependencies (npm install).
3. Run the extension.
4. Use Ctrl+Shift+Q to query the Exocortex.

⚙️ Configuration

Learning Manifest

Edit pdei_core/learning_manifest.json to add your own repositories. The system uses these to learn your specific coding patterns.

Personalities

Customize your AI's behavior by editing files in personalities/. You can define:

• Core Values: What the AI prioritizes (e.g., "IP Preservation", "Dry Humor").
• Communication Style: How the AI speaks.
• Work Cycles: Operational hours and triggers.

🤝 Contributing

Contributions are welcome. Please ensure any new modules adhere to the pdei_core/template.json structure.