CIRIS-BIO is a bio-inspired, autonomous governance platform designed to detect, contain, and evolve infrastructure defenses against critical AI system failures and "drift." It functions as an Active Immune System for AI-governed municipal and national infrastructure.
As autonomous agents are increasingly deployed within critical infrastructure—from power grids to logistics networks—traditional firewalls are insufficient. We need computational governance that operates at the speed of the systems it protects. CIRIS-BIO treats infrastructure not as a static machine, but as a living organism capable of autonomous, homeostatic responses.
- Dendritic Scanning: Real-time telemetry monitoring for pathogenic drift.
- The Fever State: Automated, localized lockdown mechanism to prevent cascading failure.
- Mnemonic Council: Autonomous governance layer utilizing Gemini-driven protocols for context-aware "Epigenetic Modifications."
- Sovereign Antibody Vault: Cryptographic archiving of defense signatures for permanent immunity against known threat vectors.
CIRIS-BIO is built upon the Sovereign AI Safety Stack (SASS):
- Frontend: Fluid-adaptive state machine using CSS Grid/Clamp for native, resolution-aware rendering.
- Governance Layer: Supabase Edge Functions integrated with Google Gemini API.
-
Telemetry: Real-time Divergence Pressure Index (DPI) calculation.
-
The governance threshold triggers intervention when:
$$D_p \times R_c > \text{Threshold}_{\text{Governance}}$$ (Where $D_p$ is Divergence Pressure and $R_c$ is Cascade Risk).
-
The governance threshold triggers intervention when:
- Dendritic Initialization: Deep scanning of infrastructure telemetry.
- Pathogenic Forecast: Identifying "drift" before an attack executes.
- Fever State Activation: Autonomous isolation of infected nodes.
- Mnemonic Council: Dynamic protocol generation via autonomous governance layer.
- Epigenetic Override: Human-in-the-loop governance gate.
- Sovereign Antibody Vault: Threat neutralization and evolutionary archiving.
/
├── index.html # Main Application Entry Point
├── tasks/ # Backend logic and task automation scripts
├── .skills/ # Adaptive governance metadata and assets
└── README.md # Documentation