VPR · Deck
03 · Research prototype
Verifiable Policy Registry
by Xcapit
So anyone can check what is blocked on school internet
Research prototype (v0.6.0): contracts not audited, no public deployment.
VPR · Deck
The problem
Protecting children online should not mean hidden censorship
Connecting a school to the internet also means protecting children from malware, scams (phishing) and harmful content, and that is done with filters. But the same filter that protects can also censor, and today nobody outside the system can check what is blocked or why.
VPR · Deck
In one picture
Like a public receipt for every block rule: the same protection, no longer in secret.
VPR · Deck
Step by step
From rule to public record, in four steps
01
The school connects and a filter protects it
The filter blocks harmful sites, such as malware or phishing pages (fake pages that steal personal data).
02
Every block rule is published
Each rule becomes a public record: what is blocked, who approved it and until when it applies.
03
Anyone can check it and appeal it
Families, teachers or auditors can review a rule and appeal it. If nobody renews a rule, it expires on its own.
04
About schools, only daily totals
Only daily totals are published about each school. Never data about a child.
VPR · Deck
A record, up close
What a public rule record would look like
Block rule · version 3
Illustrative example- Blocked category
- Phishing and malware sites
- Approved by
- 3 independent institutions · 2 signatures required
- Expires on
- 31.03.2027, unless it is signed again
- Status
- In force · open to appeal
- School data
- Daily totals only
VPR · Deck
What it offers
What changes with VPR
- 01
Protects children without hidden censorship
- 02
Anyone can check what is blocked, who decided it and until when
- 03
No data that identifies a child is ever published
VPR · Deck
Who it is for
Who it is for and what each one gains
Families
Knowing there is protection, without hidden censorship.
Schools and teachers
Seeing what their connection blocks and asking for a review when something useful for class is blocked.
Ministries and regulators
Accountability: every rule is public, signed and has an expiry date.
Auditors and civil society
Checking the rules, who signed them and their history, and filing appeals.
VPR · Deck
Under the hood
How it works under the hood
VPR separates two things: enforcement of the filtering and accountability.
01Enforcement
Done by a standard DNS resolver in the country (the service that turns a site’s name into its address).
02Accountability
Lives on an Ethereum layer-2 network (a lower-cost network built on Ethereum) that stores only cryptographic commitments: digital fingerprints of each rule, not the data itself.
VPR · Deck
Design
Design decisions, still proposed
- Proposed
Every policy version expires unless it is signed again. Proposed pilot maximum: 90 days; contract hard cap: 365 days.
- Proposed
Policies are approved by a multi-signature of independent institutions. Proposed pilot model: 2 of 3 signers.
- Proposed
Aggregates below a threshold are not published, nor is their number. Proposed starting value: k = 10. This is small-cell suppression, not differential privacy.
- Proposed
Commitments are anchored as daily Merkle roots on an Ethereum layer 2. Proposed: OP Mainnet; testnet: OP Sepolia.
Appeals store only a document hash.
VPR · Deck
The essentials
Licenses, data and project status
What anyone can verify
Anyone can check what is blocked, who decided it and for how long. Every policy expires, and any block can be appealed.
What is published about schools
Only daily aggregates with small-cell suppression. Never data that identifies a child.
Licenses
Code AGPL-3.0, schemas CC0, documentation CC BY 4.0.
Status
Research prototype v0.6.0. Contracts not audited. No public deployment. In preparation as a Digital Public Good (self-assessment).
v0.6.0 (research prototype, pre-release) · white paper v0.7 (draft for review)
VPR · Deck
Transparency
What VPR does not claim
- ×Contracts not audited
- ×No formal verification (bounded symbolic checks only)
- ×No deployment on public networks
- ×No independent review
- ×No endorsement, funding or pilot commitment from the Ethereum Foundation, UNICEF, the ITU, Giga, ministries or the DPGA
- ×Not a registered Digital Public Good
- ×ZK proofs not implemented
VPR · Deck
Roadmap
Roadmap, with a checkpoint at every stage
M0 · completed
Proof of concept (end-to-end demo, no schools)
M1
Design and approval (Gate A)
M2
Reference implementation on testnet and a proposed alpha in 2–5 schools (Gate B)
M3
Metrics, audit and mainnet registry (Gate C)
M4
Reference implementation of ZK aggregation
Phase 2 (20–50 schools) only if the gates are met and with separate funding.
School pilot: a proposal, with no agreements or commitments with any party.
VPR · Deck
For reviewers
Everything is published for review
- Repository↗ (opens in a new tab)
- Latest release (v0.6.0)↗ (opens in a new tab)
- White paper (v0.7)↗ (opens in a new tab)
- Claims and evidence register↗ (opens in a new tab)
- Guide for external reviewers↗ (opens in a new tab)
- Contract assurance (static analysis, invariants, symbolic checks, mutation testing)↗ (opens in a new tab)
- Threat model↗ (opens in a new tab)
- Privacy and re-identification analysis↗ (opens in a new tab)
- Benchmarks (includes the cost on OP Mainnet)↗ (opens in a new tab)
- Architecture decision records (ADR)↗ (opens in a new tab)
- Digital Public Good self-assessment↗ (opens in a new tab)
- Roadmap↗ (opens in a new tab)
- How to cite it↗ (opens in a new tab)
VPR · Deck
Team
Who is behind it
- Organization
- Xcapit (opens in a new tab)
- Founded in 2018 · ISO/IEC 27001
- Xcapit INC (USA) · CryptoLab S.A.S. (Argentina) · Xcapit SAC (Peru) · Córdoba · Lima · Miami
- Contact
- dpg@xcapit.com
VPR · Deck
Reviewing the project? Let's talk.
If you review projects, do research or work on school connectivity, we want your perspective.
Or write to dpg@xcapit.com
calendly.com/jose-xcapit/30min
Research prototype (v0.6.0): contracts not audited, no public deployment.