Independent researcher building Digital Biosphere Architecture for governable AI agents.
I work on a five-layer architecture for increasingly autonomous AI systems, with a focus on persona portability, semantic interaction objects, runtime governance, execution integrity, and audit evidence.
- Identifiable (persona layer)
- Interpretable for coordination (interaction layer)
- Governable (runtime control)
- Verifiable in execution (execution integrity)
- Auditable (evidence records)
flowchart LR
Persona["Persona Layer<br>POP"] --> Interaction["Interaction Layer<br>Agent Intent Protocol"]
Interaction --> Governance["Governance Layer<br>Token Governor"]
Governance --> Execution["Execution Integrity Layer<br>MVK"]
Execution --> Audit["Audit Evidence Layer<br>ARO-Audit"]
- digital-biosphere-architecture - architecture hub and narrative entry point
- verifiable-agent-demo - smallest end-to-end walkthrough across layers
- agent-intent-protocol - interaction-layer draft for intent, action, and result objects
Portable persona objects for the Persona Layer.
Machine-readable intent, action, and result objects for agent runtimes.
Runtime governance and checkpoint control for the Governance Layer.
Execution integrity and verification-oriented runtime truth for the Execution Integrity Layer.
Evidence records, receipts, and reviewable exports for the Audit Evidence Layer.
Minimal cross-layer demonstration that links persona, interaction, governance, execution trace, and audit evidence.
| Layer | Repository |
|---|---|
| Persona Layer | persona-object-protocol |
| Interaction Layer | agent-intent-protocol |
| Governance Layer | token-governor |
| Execution Integrity Layer | fdo-kernel-mvk |
| Audit Evidence Layer | aro-audit |
| Cross-layer demo | verifiable-agent-demo |
This work is not aimed at replacing existing agent frameworks. The focus is on governance-oriented architecture layers that can be attached to AI systems as reusable, inspectable, and standardizable components.
- Agent Interaction Protocols
- Semantic Object Communication for AI agents
- Runtime governance for autonomous AI systems
- Execution integrity and bounded audit evidence
- Digital Biosphere Architecture - canonical architecture hub
- Persona Object Protocol - persona-layer entry point
- Agent Intent Protocol - interaction-layer draft
- Verifiable Agent Demo - compact cross-layer walkthrough