Skip to content

Add Nobulex: Trust Capital / behavioral evidence layer for AI agents #32

Description

@arian-gogani

Nobulex fills a gap in the current list: tamper-evident behavioral evidence that accumulates into earned reputation.

Every agent action produces a bilateral Ed25519 receipt (one signature before execution binding the authorized scope, one after binding the actual outcome), hash-chained so the full history is verifiable by any third party.

Over time those receipts accumulate into Trust Capital, a machine reputation that gates what agents are allowed to do. New agents start restricted. Proven agents earn broader access.

Where it fits in this list:

Could go under Governance/Compliance alongside NeMo Guardrails and Guardrails, or as a new section for Behavioral Evidence / Audit Trail tools. The key difference from the existing categories: Nobulex does not block or filter. It proves what happened and converts that proof into economic access.

Details:

  • Repo: https://github.com/arian-gogani/nobulex
    • License: MIT
      • Stack: TypeScript, Ed25519, JCS/RFC 8785 canonicalization
        • Microsoft merged the receipt primitive into their Agent Governance Toolkit (PRs #1302, #1333)- Four independent implementations cross-validated (byte-match verified)- Under AAIF (Linux Foundation) staff review- 5 npm packages, 24 GitHub starsHappy to submit a PR if you point me to the right section.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions