Video thumbnail

VeriKnowledgeOnchain

Decentralized platform for verifiable onchain knowledge. Purchaseable data stored securely with encrypted shareable access via Lit Protocol. Dynamic knowledge graph learns from real-time interactions, powering metaverse bots and assistants.

VeriKnowledgeOnchain

Created At

Agentic Ethereum

Project Description

Data Storage & Verification: Knowledge is tokenized and stored on the blockchain, ensuring immutability and traceability. Smart contracts govern data ownership, usage rights, and transaction processing for purchaseable content.

Secure Access: Data is encrypted and access is managed via the Lit Protocol, allowing for secure, shareable, and permissioned decryption. This means that only authorized users or agents can access sensitive or premium knowledge content.

Dynamic Knowledge Graph: The system builds a dynamic, context-rich knowledge graph that updates in real-time. Each data node is linked to provide contextual relevance and is designed to grow from user interactions. This evolving structure allows the graph to reflect new insights and data relationships.

Interactive Agents: The knowledge graph integrates with AI-driven agents, such as metaverse companion bots and assistants. These agents utilize the graph for context-aware interactions and can learn over time, enhancing user experiences with up-to-date, verifiable information.

Enhancement with Advanced Technologies: Retrieval Augmented Generation (RAG) and semantic search algorithms like Deepseek are used to extract and present the most relevant information from the knowledge graph. Additionally, DeFi components facilitate seamless, automated transactions, enabling users to purchase and exchange knowledge data securely.

How it's Made

Langchain/Langgraph: We use Langchain to process natural language inputs and generate dynamic queries. Langgraph then structures these into a coherent knowledge graph, linking data nodes based on contextual relationships.

Neo4j: For managing and querying the knowledge graph, we rely on Neo4j, a graph database that efficiently stores relationships and supports complex queries. It forms the backbone of our dynamic data structure.

Walrus: We integrated Walrus for robust encryption of our data. This tool allows us to securely encrypt and decrypt knowledge, ensuring only authorized parties access sensitive information. Our approach was a bit hacky: we customized Walrus to work seamlessly with our decentralized key management, optimizing access speeds without compromising security.

Privy: To bolster privacy, we implemented Privy. This tool ensures that all user data and transactions adhere to strict privacy standards, and it simplifies compliance by automating privacy controls and notifications.

By combining these technologies, we created a modular system where each component plays a crucial role. Partner technologies like Neo4j provided proven frameworks that boosted our development speed and reliability. Our innovative use of Walrus for encryption and the tailored integration with decentralized key management stands out as a particularly hacky, yet effective, solution that ensured both security and performance.

background image mobile

Join the mailing list

Get the latest news and updates