EthosIntel

Problem Statement

Our project is a decentralized, secure, and privacy-preserving platform for digital evidence management. We leverage the Flow blockchain for its speed and developer-friendly features, the Lighthouse SDK for immutable and encrypted storage, and the ASI AgentVerse for autonomous AI analysis.The process begins when an investigator or a submitter uses a client-side application to upload evidence. This file is encrypted and stored using the Lighthouse SDK, which provides permanent, decentralized storage on IPFS/Filecoin. The SDK returns a unique CID (Content Identifier) for the encrypted file.A transaction is then submitted to a smart contract on the Flow blockchain. This transaction records the CID and a cryptographic hash of the original, unencrypted file, creating a public, tamper-proof record of the evidence.This on-chain event triggers an autonomous AI agent deployed on the ASI AgentVerse. This agent, with its own verifiable identity, analyzes the evidence. It can't decrypt the file but can process the metadata and interact with other data sources to provide a structured report. This report is also uploaded to Lighthouse, and its CID is linked back to the original evidence record on the blockchain.The result is a complete, end-to-end flow where data is encrypted from the start, its integrity is verified on-chain, and an autonomous AI agent provides an immutable, transparent analysis.

Solution

We built this project by tightly integrating a few key technologies to create a seamless, end-to-end workflow, focusing exclusively on the Flow, ASI, and Lighthouse ecosystems.Technologies and Integration Decentralized Storage (Lighthouse SDK): The Lighthouse SDK is the foundation of our storage layer. We used it to handle the entire file management process, from encrypting the evidence client-side to uploading it to decentralized storage. This provides a secure and permanent location for the data, with the unique CID serving as the on-chain pointer.Blockchain (Flow): We developed a smart contract on the Flow blockchain to serve as our immutable ledger. It's designed to be lightweight, storing only the CIDs and hashes. We utilized the Flow CLI to compile and deploy the contract and the Flow SDK to enable interaction from our client-side and backend services. Flow's high throughput and low fees are ideal for this kind of data-logging application.AI Agent Layer (ASI AgentVerse): The core intelligence of our system is an autonomous AI agent running on the ASI AgentVerse. This agent is programmed to listen for new evidence submission events on the Flow blockchain. When a new event is detected, the agent automatically retrieves the evidence from Lighthouse using the provided CID. While it cannot decrypt the private file, it can analyze any publicly available metadata and use its reasoning capabilities to produce a structured report. This report is then uploaded back to Lighthouse, and the agent calls a smart contract function to link it to the original evidence.Notable Aspects The most "hacky" or notable aspect of this project is the autonomous, trustless orchestration between the blockchain and the AI agent. Instead of relying on a centralized server to manage the workflow, the system is designed to be fully decentralized. The Flow blockchain acts as a public signaling layer, triggering the ASI AgentVerse agent to perform its analysis. The agent operates independently and its report is then immutably recorded, creating a truly transparent and decentralized pipeline.