Framework demonstrates machine-verifiable protocol compliance in AI systems, suggesting improved interoperability.
This work introduces the Agent Object Protocol (AOP), a framework for machine-adjudicable protocol compliance based on executable artifacts rather than textual interpretation. Traditional protocols rely on human interpretation of specification text, which often leads to ambiguity across implementations. AOP reframes protocol interoperability as a reproducible adjudication process over executable artifacts, including schemas, conformance fixtures, and CI validation gates. The repository accompanying this work provides a reproducible artifact baseline including: • protocol schemas • conformance fixtures • adversarial invalid tests • CI adjudication gates • reproducible evaluation logs Artifact baseline: Repository: https://github.com/joy7758/agent-object-protocol Code baseline: main@c9f94ee Evidence assets: 41242b1 CI evidence:https://github.com/joy7758/agent-object-protocol/actions/runs/22717328360 Figure S8 demonstrates deterministic rejection of adversarial invalid fixtures (total=3; parse_reject=1; checks_reject=2; unexpected_pass=0). This work proposes a methodological shift toward Protocol-by-Artifacts, enabling machine-verifiable protocol adjudication for AI artifact ecosystems and software supply-chain interoperability.
No takes yet. Share an insight, caveat, or question.
Bin Zhang (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: