The Interoperability Problem No Product Passport Platform Solves
Why supply-chain transparency needs a decentralized agent network — and what that means for manufacturers
The Gap Between Having Data and Making It Useful
Most manufacturers already have the data that regulators, customers, and trading partners want. It exists — scattered across ERP systems, PLM tools, MES platforms, certification databases, supplier portals, and internal documents. The problem is not data scarcity. It is data fragmentation.
A battery manufacturer can produce a chemical composition report, a carbon footprint assessment, a chain-of-custody record, and a conformity certificate from an accredited body. But getting those four artifacts into a form that a downstream customer's procurement system, a regulator's risk-scoring engine, or a recycler's material-recovery process can actually consume — that is the hard part. It requires mapping fields between incompatible systems, establishing trust between organizations that have no prior relationship, and maintaining the integrity of evidence as it changes hands across jurisdictional boundaries.
This is the interoperability problem. And it is the reason most digital product passport initiatives stall after the pilot phase.
Why Point-to-Point Integration Fails at Scale
The conventional approach to supply-chain data exchange is point-to-point integration. Manufacturer A builds a custom API connection to Customer B. Customer B builds another connection to Regulator C. Regulator C builds yet another to Auditor D. The number of connections grows quadratically with the number of participants. Each connection is a maintenance burden, a security surface, and a potential point of failure.
More fundamentally, point-to-point integration assumes that the parties know about each other in advance. It breaks the moment a new downstream participant — a recycler in a different country, a customs authority using a different system, a consumer scanning a QR code — needs to verify a claim. There is no pre-arranged connection. There is no shared data model. There is no common trust framework.
What is needed is not another platform that aggregates data in a central repository. That model creates a honeypot of commercially sensitive information, a single point of failure, and a governance nightmare. What is needed is a way for independent organizations to issue, discover, and verify claims about products and supply chains without requiring a central coordinator — and without requiring every participant to replace the systems they already run.
A Decentralized Approach: The UNTP Foundation
The United Nations Transparency Protocol (UNTP), developed under UN/CEFACT, provides the standards foundation for this approach. UNTP defines a set of verifiable credential types — product passports, facility records, traceability events, conformity credentials — that can be issued by any organization and verified by any other, without a central authority.
The key insight is that UNTP credentials are self-contained units of trust. A digital product passport issued by a manufacturer carries its own cryptographic proof of integrity, its own schema definition, and its own access-control metadata. A verifier does not need to call the issuer's database to check whether the credential is valid. The credential itself, combined with the issuer's public DID document and a status list, is sufficient.
This is a profound departure from traditional supply-chain data exchange. It means that a product passport can travel with a physical product — on a QR code, in an NFC tag, in a shipping document — and be verified by any party, including parties that were unknown when the credential was issued. No prior relationship required. No bilateral integration required.
From Specifications to an Operational Network
But specifications alone do not operationalize a network. UNTP defines what verifiable supply-chain data looks like. It does not define how independent organizations, each running different enterprise systems, continuously issue, discover, verify, and update that data at scale.
This is the gap that Ourbits AgentBridge is designed to address. AgentBridge is not a platform that stores supply-chain data. It is an interoperability layer — a set of specialized software agents that embody UNTP roles, communicate using verifiable-credential messages, and collaborate across organizational boundaries.
The distinction matters. A platform aggregates data and controls access to it. An agent network distributes intelligence and leaves data where it belongs — with the organizations that created it. Each participant runs their own agents, controls their own data, and decides what to share, with whom, and under what conditions. The network provides the protocols, the schemas, and the trust framework. The participants provide the data and the authority.
The Architecture in Brief
AgentBridge is organized around a small number of core design principles:
No central coordinator. The system has no mandatory central services beyond the open, UN-maintained registers that are already part of the UNTP governance model. A manufacturer can issue product passports using only their own infrastructure. Downstream parties discover those passports later through standard identity resolution.
Verifiability by unknown parties. Every credential and every agent message is cryptographically protected. A verifier does not need to trust the party that forwarded the data — they verify the credential's signature, check its status against a public list, and confirm the issuer's real-world identity through a digital identity anchor. Trust is established cryptographically, not through a central access-control server.
Support for any maturity level. Not every supply-chain participant runs a sophisticated DID infrastructure. The architecture supports simpler participants through bridge agents that translate between enterprise systems and the agent network. A small supplier using a spreadsheet-based inventory system can participate through a bridge that maps their data to UNTP credentials — without replacing their existing tools.
Selective disclosure. Supply-chain transparency does not mean making all data public. The architecture supports multiple access-control patterns, from fully public product summaries to encrypted credentials accessible only to authorized role-holders. A single product passport can simultaneously serve a consumer scanning a QR code, a regulator with appropriate credentials, and a competitor who should see nothing — all enforced through cryptographic policy, not through a central gatekeeper.
Extensibility without fragmentation. New regulatory requirements — battery passports, deforestation due-diligence, textile sustainability — are absorbed through a versioned schema knowledge base and a conformity vocabulary catalog, not through changes to the core agent codebase. Sectoral communities can publish profiles that extend the base schemas without breaking global interoperability.
What This Means in Practice
Consider a battery manufacturer preparing for the EU Battery Regulation. They need to issue a digital product passport for each battery placed on the market, linking it to material provenance, carbon footprint, and conformity evidence from independent assessment bodies. Downstream, an automotive OEM needs to verify those passports as part of their own due-diligence process. A recycler at end-of-life needs to access material composition data to optimize recovery. A regulator needs to audit the chain of custody.
In a point-to-point world, the manufacturer builds separate integrations for each of these participants. In a centralized platform world, they upload their data to a shared repository and hope the platform's access controls are sufficient.
In the agent network world, the manufacturer issues a verifiable product passport. The OEM discovers it through identity resolution and verifies it independently. The recycler accesses the material composition data through a shared-secret mechanism authorized by the manufacturer. The regulator verifies the full chain of custody using their role credentials. No bilateral integrations. No central data repository. No single point of failure.
The manufacturer retains control of their data at all times. They can revoke access, update the passport as new information becomes available, and prove exactly what was shared with whom through an auditable log of credential exchanges.
The Broader Implications
This architecture has implications beyond any single regulation or industry.
For manufacturers, it reduces the integration burden of compliance. Rather than building custom connections to every downstream partner, they run agents that speak a common protocol. The agents handle schema validation, credential issuance, and discovery registration.
For regulators and customs authorities, it enables machine-verifiable compliance. A cross-border compliance pack can be assembled from verifiable credentials issued by independent participants — without requiring those participants to submit data to a central registry.
For conformity assessment bodies, it provides digitally native certificates that travel with products. A conformity credential issued by an accredited body is discovered and verified by any downstream agent, reducing the cost and friction of conformity evidence exchange.
For software and solution providers, it offers a clear interoperability target. Rather than building proprietary data models, providers can build UNTP-native modules that participate in the broader network.
For the circular economy, it enables post-sale lifecycle tracking. A product's passport remains alive after sale, accumulating verified lifecycle events — repairs, resale, recycling — that support accurate end-of-life accounting and secondary-material markets.
The Path Forward
The technical foundations are in place. The UNTP specifications are maturing toward a 1.0 release. The verifiable credentials ecosystem — W3C VCDM 2.0, DID Core, Bitstring Status Lists — is production-ready. The remaining work is implementation: standing up the schema knowledge base, building reference agents, developing enterprise bridges, and running multi-organization pilots.
The goal is not another closed platform that locks supply-chain data into a proprietary silo. It is infrastructure that allows product, facility, conformity, and lifecycle data to move between systems with less custom integration and stronger verification — a decentralized network for verifiable supply-chain transparency.
The interoperability problem is not a data problem. It is a trust and coordination problem. And it will not be solved by building bigger platforms. It will be solved by building better protocols — and by giving those protocols an operational layer that can discover, verify, and act on behalf of the organizations that depend on them.
This article is based on publicly available UNTP specifications and the Ourbits AgentBridge architectural vision. The system is currently in design and pilot preparation.
