Show HN: I'm 15 and built a cryptographic accountability layer for AI agents (github.com)

A 15-year-old developer has unveiled Nobulex, a cryptographic accountability layer designed to ensure AI agents adhere to their operational rules. This protocol allows agents to declare behavioral covenants, enforce them at runtime, and generate a tamper-evident audit trail of their actions. For instance, if an agent commits to only transferring amounts under $500, the system checks every action against these rules before allowing execution. If tampering occurs, as shown by a third agent with a broken hash chain, transactions are automatically blocked, ensuring trust and compliance. Nobulex is significant for the AI/ML community as it bridges a crucial gap in agent governance: the lack of proof that agents follow their established policies at the action layer. The protocol employs three main primitives—Declare, Enforce, and Prove—to deliver deterministic, efficient, and mathematically verifiable behavior without relying on complex machine learning or heuristics. By providing an independent verification mechanism for agents, Nobulex fosters trust in automated systems, promoting safer interactions in environments where autonomous decision-making is critical. This innovative framework could lead to wider adoption of AI agents across various sectors, enhancing accountability and reliability in AI-driven processes.