GitHub Repository

The trust economy for autonomous AI agents. Credit scores for machines. Agents earn Trust Capital through verified behavior, gating what they are allowed to do. Autonomy earned, not granted.

31 starsTypeScript

I'm 15. I mass published 134K lines to hold AI agents accountable

Name: I'm 15. I mass published 134K lines to hold AI agents accountable
Availability: InStock
Author: nobulexdev

by nobulexdev·Mar 1, 2026·6 points·23 comments

Visit Project View on HN

AI Analysis

●●●BangerZero to OneBig Brain

Cedar-inspired DSL + W3C DID + hash-chained logs verify agent compliance deterministically.

Strengths

•Core insight is genuinely clever: audit actions against covenants, not the neural network itself—always decidable
•Six-primitive architecture (identity, covenant, attestation, log, verification, enforcement) is cohesive and well-structured
•Deterministic verification with violation proofs means you can prove breach without trusting the agent or platform

Weaknesses

•No on-chain enforcement deployed yet; enforcement tier requires staking contracts that aren't live
•Extremely early; no production users or case studies demonstrating whether covenants solve real trust issues in practice