Open standard for honest product discovery based on verified data
An open standard for honest product discovery.
"You don't need an opinion about a product. You need its data."
The internet was supposed to make choosing easier. It made it worse.
Search results are auctions. Reviews are manipulated. Recommendations are paid placements. Every platform that started as a meritocracy became a marketplace for visibility — where the biggest budget wins, not the best product.
The user asks a question. They get marketing.
DPPchain is a response to this. Not another platform. Not another algorithm. A standard — open, forkable, ungovernable by any single entity — that lets AI agents answer product questions using verified data instead of curated content.
No reviews. No sponsored results. No visibility auctions.
Facts. Compared against your needs. With a full audit trail of where every fact came from.
The EU Digital Product Passport (DPP) regulation mandates that products carry structured, machine-readable data about their composition, origin, sustainability, technical parameters, and supply chain. Starting 2026 with textiles and electronics, expanding across all major product categories.
For the first time in history, product data has:
- Legal accountability — falsifying a DPP is a civil and criminal liability
- Standardized structure — machine-readable across manufacturers and markets
- Verified origin — issued through certified bodies, not self-reported marketing
This is the data layer DPPchain is built on.
DPPchain is an open protocol for AI agents that:
- Ingest structured product data from DPP registries, technical datasheets, certification bodies, and official standards databases
- Cross-verify data across independent sources — inconsistencies are flagged, not smoothed over
- Match verified product attributes against user-defined preferences and requirements
- Return a recommendation with a complete, human-readable audit trail: every claim cites its source
The agent does not visit product pages. It does not read reviews. It does not know who has the bigger advertising budget.
It reads data. It compares data. It tells you what fits.
1. Verified sources only The agent operates exclusively on data with legal accountability behind it. DPP registries, CE certification databases, ECHA chemical records, ISO-compliant technical sheets. No self-reported marketing data.
2. Full source transparency Every recommendation comes with a complete audit trail. The user can inspect exactly which data points led to which conclusion. The reasoning is never a black box.
3. Zero monetization of visibility No entity can pay for a higher position in results. The only ranking criterion is fit to user requirements. This principle is structural, not a policy — it is enforced by the architecture.
4. User preferences are sovereign Preference profiles are defined by the user, stored locally, and never shared with third parties or used for targeting. The personalization layer belongs to the person, not the platform.
5. Open source without exceptions The protocol specification, reference implementation, and all core tooling are and will remain fully open source. Anyone can implement, fork, extend, or audit. No proprietary lock-in, ever.
The person buying — who wants to know if the jacket actually contains what the label claims, whether the laptop's battery specs are real, whether the building material meets the standard it's certified for. Not what the brand says. What the data says.
The company selling honestly — that has a genuinely good product and keeps losing to competitors with larger ad budgets and better SEO agencies. DPPchain cannot be gamed by spend. It can only be won by product quality reflected in verifiable data.
The developer building — who wants to create tools that help people make better decisions, and now has a trustworthy, structured, legally-accountable data layer to build on.
Every system built on reviews, ratings, and recommendations has one structural weakness: opinions can be manufactured.
Five-star reviews are bought. Influencer recommendations are paid. "Editor's choice" badges are negotiated. Link-building agencies exist because authority can be simulated.
DPPchain does not use opinions. A product either has a certified recycled cotton content of 80% or it doesn't. A motor either produces 150kW or it doesn't. A material either meets EN 13501 fire resistance classification or it doesn't.
When data has legal weight behind it, the cost of falsification becomes a deterrent. DPPchain is built on the layer of the internet where lying has consequences.
Early specification stage.
The DPP regulatory framework is coming into force. The data infrastructure is being built by industry. The missing piece is the agent layer — the protocol that lets a user's AI assistant consume this data and use it for honest recommendation.
That is what DPPchain specifies.
This project needs people who understand:
- DPP technical specifications and EU regulatory landscape
- AI agent architecture and tool-use patterns
- Data schema design and cross-registry normalization
- The problem this solves, personally or professionally
If you've read this far and something clicked — open an issue, start a discussion, or reach out directly.
The standard will be shaped by the people who build it first.
MIT. Take it, use it, build on it, improve it. The only thing you cannot do is close it.
DPPchain is a working name. The protocol will be named by the community that builds it.