Skip to content

zico-hubb/auditAlly

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 

Repository files navigation

Audit Ally

Decentralized Forensic Integrity & Agentic Protocol Verification System

Fetch.ai ASI AI Stack

Audit Ally is an Omni-Domain forensic auditing ecosystem that eliminates the single point of failure in legal chain-of-custody protocols. By combining Fetch.ai uAgents, deterministic ASI-1 Reasoning, and Web3 Immutable Ledgers, Audit Ally ensures that physical and digital evidence remains scientifically valid and legally admissible.


The Problem

Centralized evidence logs are vulnerable to human error, chronological paradoxes, and retroactive tampering. Furthermore, biological evidence (like DNA meant for STR profiling) or volatile evidence (like arson VOCs) rely on physical environments that are rarely cross-referenced with cryptographic certainty.

The Solution

Audit Ally introduces a Zero-Trust Forensic Pipeline:

  1. uAgent IoT Witnesses: Physical evidence lockers stream live telemetry (temperature, VOCs, RFID) via autonomous digital twins.
  2. Agentic Auditing (ASI-1): An AI iteratively evaluates human-written logs against strict ISO/IEC 27037 standards and live IoT telemetry.
  3. Immutable Anchoring: The AI's reasoning trace and the source evidence hash are permanently signed to the Fetch.ai/ASI Dorado Testnet.

Highlights

1. Highly Structured Prompting

Audit Ally avoids standard conversational LLM wrappers. The ASI-1 model is constrained by highly engineered, XML-tagged structural prompts. It is forced to evaluate inputs against strict, deterministic scientific standards (e.g., temporal continuity, digital forensic acquisition protocols, and biological degradation thresholds).

2. Multi-Step Agentic Iteration

The system employs an autonomous "Red-Team" loop:

  • Iteration 1 (The Scan): ASI-1 performs a comprehensive multi-domain scan of the ingested PDF.
  • Iteration 2 (The Critique): ASI-1 is forced to play Devil's Advocate against its own findings, actively searching for benign environmental or latency-based explanations before finalizing a JSON-formatted risk score.

3. uAgent IoT Integration

Audit Ally features a dedicated live connection to a Fetch.ai uAgent swarm (agent1qvk...haki88). This allows the ASI-1 model to dynamically reason on live hardware telemetry, bridging the gap between the physical environment and the decentralized ledger.


System Architecture

  • Frontend: React.js (High-Fidelity Dashboard, Live Telemetry Visuals, PDF Generation).
  • Backend Brain: Python / Flask.
  • Agentic Framework: Fetch.ai uAgents library for IoT simulation.
  • Blockchain SDK: cosmpy for automated transaction signing and payload anchoring to the Dorado Testnet.
  • Smart Verification: Cryptographic SHA-256 hashing of both source files and AI logic traces.

Local Setup & Deployment

Prerequisites

  • Python 3.9+
  • Node.js & npm
  • A funded Fetch.ai wallet (Dorado Testnet)

1. Start the Backend & AI Engine

cd backend
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install -r requirements.txt
python app.pys

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors