Can this skill be used for security auditing?

Absolutely. It is specifically designed for high-integrity behavioral verification and active threat detection in automated coding and development environments.

What happens when a prediction doesn't match the observation?

The Comparator calculates an error signal magnitude. If the mismatch exceeds defined thresholds, the system amplifies the signal and triggers a WARNING or CRITICAL threat alert.

What is the main purpose of the Reafference & Corollary Discharge skill?

It uses neuroscience principles to help Claude distinguish between its own generated actions and external signals, filtering out 'self-noise' to identify genuine anomalies.

How does the system predict interaction outcomes?

It uses SHA-256 hashing of interaction content (like file paths or descriptions) to create a deterministic 'Efference Copy' represented as a color index.

Does this skill require a database for operation?

Yes, it integrates with DuckDB to store and analyze efferent commands, sensory reafference matching scores, and suppressed or amplified signals across sessions.

Behavioral Verification & Reafference

Name: Behavioral Verification & Reafference
Author: plurigrid

byplurigrid

•

Security & Testing

Implements neuroscience-based reafference theory to distinguish self-generated AI actions from external anomalies and threats.

About

This advanced skill applies von Holst’s reafference principle to AI interaction analysis, allowing Claude to predict the sensory feedback of its own actions and cancel out 'self-noise.' By comparing deterministic predictions (Efference Copies) against actual interaction history (Sensory Reafference), the system can precisely identify mismatches as external threats or anomalies. This provides a robust framework for behavioral verification, ensuring high-integrity signal processing and automated threat escalation within development environments.

Key Features

Real-time Sensory Reafference matching against Claude's interaction history
2 GitHub stars
Deterministic Efference Copy generation using SHA-256 hashing for action prediction
Comprehensive DuckDB integration for persistent tracking and temporal analysis of signals
Automated Comparator logic for error signal computation and threat level classification
Corollary Discharge mechanism for autonomous signal suppression or amplification

Use Cases

Verifying the integrity and consistency of long-running automated development tasks
Implementing advanced anomaly detection in AI-driven workflows based on neural mechanisms
Distinguishing between intended AI-generated outputs and unauthorized external modifications

About

Key Features

Real-time Sensory Reafference matching against Claude's interaction history
2 GitHub stars
Deterministic Efference Copy generation using SHA-256 hashing for action prediction
Comprehensive DuckDB integration for persistent tracking and temporal analysis of signals
Automated Comparator logic for error signal computation and threat level classification
Corollary Discharge mechanism for autonomous signal suppression or amplification

Use Cases

Verifying the integrity and consistency of long-running automated development tasks
Implementing advanced anomaly detection in AI-driven workflows based on neural mechanisms
Distinguishing between intended AI-generated outputs and unauthorized external modifications