What the study found
The essay argues that in agent-attestation frameworks, adding a central registry for agents, keys, compliance badges, search, reputation, or verification-as-a-service can quietly turn an offline-verifiable trust framework into a platform. It recommends a distributed-trust pattern instead.
Why the authors say this matters
The authors say this matters because registry vocabulary can change the nature of the framework itself. They conclude that the alternative they recommend uses public key-history mirrors, operator-managed trust anchors, conformance vectors rather than certification, and no framework-level discovery layer.
What the researchers tested
The essay uses the Agent Evidence Package (AEP) profile and EATF reference verifiers as a worked case to analyze the architectural choice between marketplace-style design and distributed trust. In version v0.4, it also adds a related-work paragraph, expands the surveillance-surface argument with examples from OpenPGP keyserver, OCSP, package-registry, Certificate Transparency, and Rekor, and adds a submission strategy.
What worked and what didn't
The essay presents the distributed-trust pattern as the recommended alternative. It describes the registry-centered approach as one that can convert the framework into a platform, and it contrasts that with public key-history mirrors, operator-managed trust anchors, conformance vectors, and no framework-level discovery layer.
What to keep in mind
This is an essay draft, identified as publication draft v0.4, so the available summary reflects that version. The abstract does not describe empirical testing results, and it does not provide limitations beyond the scope and examples listed. Tyche Institute is stated to be a research entity, not a trust service provider, and EATF is described as an open specification and reference implementation, not as an eIDAS trust service or compliance certification product.
Key points
- The essay argues that central registries can turn agent-attestation frameworks into platforms.
- It recommends a distributed-trust pattern instead of framework-level discovery and registry features.
- The worked case is the Agent Evidence Package (AEP) profile and EATF reference verifiers.
- Version v0.4 adds related work, surveillance-surface examples, and a submission strategy.
- The abstract does not report empirical test results or separate study limitations.
Disclosure
- Research title:
- Registry choices can turn agent trust frameworks into platforms
Get the weekly research newsletter
Stay current with peer-reviewed research without reading academic papers — one filtered digest, every Friday.