Grant Application SHPH: High-Assurance Post-Quantum Network Layer for Censorship Resistance

Community Grants Applications
1

Hi Zcash Community and ZCG Committee (@zcash-community-grants),

We have submitted an official grant application for SHPH (Shroud-Phantom).

Project summary

SHPH is an open-source encrypted transport and overlay written in Rust. It is designed to provide an additional connectivity option for privacy-preserving applications and infrastructure operating on unreliable or heavily restricted networks.

SHPH is not an anonymity network and is not intended to replace Tor, Nym, WireGuard, or existing Zcash privacy technology. It currently remains a developer-stage project for controlled laboratory evaluation; it has not received an independent security audit and should not yet be considered production-ready.

Current status: v0.6.4-dev.4

The latest development release includes:

  • Hybrid ML-KEM-768 and X25519 key establishment with transcript-bound key derivation.
  • Memory-safe Rust implementation with explicit key-material cleanup.
  • Linux native-TUN and Windows Wintun integration paths.
  • Fail-closed route and underlay-bypass handling, tested across Windows and Linux environments.
  • Configurable Shroud traffic-shaping profiles with discrete size padding and handshake-length variation.
  • SOCKS5 underlay support for deployments where direct IP connectivity is unavailable.
  • Simplified host and join workflows, shareable tickets, identity inspection, and diagnostic commands.
  • Reproducible fuzzing, validation, Criterion benchmarks, and cross-platform CI artifacts.

Our current benchmarks are useful engineering evidence, but they are not claims of real-world censorship resistance. They are primarily local, deterministic, or CI-hosted measurements. Results show strong replay-validation and handshake performance, while also showing that padding overhead remains significant for small-packet traffic such as SSH and VoIP. We are documenting those limitations rather than presenting the best-case results alone.

Why this may be relevant to Zcash

Zcash already benefits from established privacy and censorship-resistance tools, including Tor. SHPH would be an optional complementary transport rather than a replacement.

The practical use case we want to validate is a deployable additional connectivity path for Zcash-related infrastructure, such as:

  • lightwalletd operators;
  • full-node operators;
  • wallet and infrastructure developers;
  • community services operating in networks where direct connectivity is unreliable or blocked.

The grant would not be used simply to produce more general-purpose VPN features. Its purpose would be to harden SHPH, independently review its design and implementation, and demonstrate a concrete Zcash-oriented integration or pilot. The final deliverables should make it possible for the community to evaluate whether SHPH provides meaningful incremental value alongside Tor.

Proposed milestones

Milestone 1 — $12,000

Target completion: October 15, 2026

  • Harden the transport, framing, configuration, and control-plane code.
  • Expand continuous fuzzing and regression testing.
  • Complete interface-scoped NAT and route-cleanup safeguards.
  • Improve Windows and Linux deployment reliability.
  • Publish reproducible benchmark methodology and results, including limitations.
  • Produce a documented Zcash integration prototype or test deployment.

Milestone 2 — $13,000

Target completion: December 15, 2026

  • Complete an adaptive Shroud profile implementation, subject to security and performance review.
  • Evaluate multi-hop or alternate-underlay support only where it directly improves the Zcash deployment case.
  • Conduct an external technical review focused on protocol design, key handling, framing, replay protection, and routing behavior.
  • Publish deployment documentation and an operator-focused Zcash pilot report.
  • Release the resulting implementation and evidence publicly.

The dates are targets, and the scope may be refined with ZCG and community feedback. We would prefer to deliver a smaller, independently reviewable Zcash integration than overstate the maturity or usefulness of the project.

We welcome technical questions, architectural criticism, and feedback on whether this represents a useful addition to the existing Zcash privacy and connectivity ecosystem. ( This is a rewrite from the past application )

1 Like
Zcash Community Grants Meeting Minutes 8/31/2026
2

When someone copy-pasted AI output with so little oversight they use the wrong username and blatantly pasted Markdown… my god.

The second question being five words: Why Not Tor or Nym?

Tor was built by veteran researchers at the US Naval Research Laboratory to protect US spies and intelligence, at astronomical stakes, bootstrapped with near infinite funding compared to $25,000, matured for over three decades with over 100 core developers working on it full time, with their own foundation and grants.

You… you want $25k to build an alternative to THAT. Not proof of concept. “General availability” and deployment? Just… what?

EDIT:

Russia’s premier foreign intelligence agency (SVR) uses Tor to host their confidential tipline. Even they believe in it.

3

Hello mate, sorry, I use ai for my summaries and also my documents, I can assure you that the quality isn’t going to miss out, also if you encounter any problems, definitely tell me, also sorry for my bad English.

4

Also if you check our documents, we have always stated that we aren’t going to rival wireguard ( because of its years of experience and also battle tested, OpenVPN because it’s just the pioneer ), also you said Tor, Tor is a whole different story, we can never rival them, but as I said, we try our best.

5

Also on your Edit, I didn’t say that russia doesn’t use Tor, if you count Russia in for using Tor, you might just count England’s Mi6 for using Tor for their whistleblowers or informers.

6

I- I did. I counted in US (their spies), Russia (their spies). My point was that if Tor is secure enough for literal spies, why is this more beneficial than using Tor?

7

Also I think we’re getting a little crossed up here, so let me clarify what I actually mean. I’m not claiming SHPH can replace Tor, Nym, WireGuard, or any of the other projects mentioned. That isn’t the goal of the project, and I don’t think it would be realistic to make that claim.

SHPH is targeting a different problem and a different layer: a hardened, auditable encrypted transport that can operate under hostile network conditions, with post-quantum key establishment, traffic-shaping/countermeasure mechanisms, endpoint isolation, and VPN/TUN integration.

The $25k request isn’t “$25k to build a Tor replacement.” It’s funding for the specific engineering and verification milestones in the proposal: hardening the existing implementation, independent testing/fuzzing, improving the censorship-resistance mechanisms, and producing reproducible evidence around those claims.

I also want to be clear that I’m not presenting SHPH as something that’s already equivalent to a mature production network like Tor. If some wording in the application gave that impression, that’s something I can clarify and improve.

I genuinely appreciate the criticism. SHPH is still a young project, and I’d rather have people challenge the architecture and claims now than have those questions ignored later.

I’m happy to discuss the technical differences and limitations in good faith. I’m not trying to argue that SHPH is “better than Tor”; I’m trying to build something that solves a different problem and can potentially complement existing tools. Best of regards.

1 Like
8

Tor is:

  • Hardened
  • Audited over a hundred times independently
  • Encrypted
  • A transport (globally available with over 8,000 exit nodes and over a million nodes at times
  • A post-quantum migration underway
  • Can operate in the most hostile network conditions
  • Has traffic-shaping and countermeasures, packages into blocks
  • Endpoint isolation
  • Literally the most famous use case of SOCKS5 and SOCKS5H

The issue is that you’re on Zcash Community Grants. Why is your project relevant FOR ZCASH USERS, when most wallets already implement Tor or Nym?

If this was AnonymityCon 2026 Grants then you’d be 100% right to apply.

9

I think there are a couple of important points I didn’t explain clearly enough in my previous reply.

First, self-hosting is a deliberate part of the design. SHPH isn’t intended to depend on a single operator, commercial VPN provider, or centralized service. An individual, organization, wallet provider, or Zcash infrastructure operator can run their own transport infrastructure and control the deployment themselves. That matters in censorship environments where the infrastructure providing connectivity can itself become a point of failure.

Second, SHPH is open source and intended to remain modifiable and auditable. The point isn’t just that people can inspect the code; operators can adapt the transport to their own requirements, deploy their own instances, and integrate it into other software without depending on SHPH’s authors as a service provider. I’m deliberately trying to build a transport primitive rather than another centralized VPN service.

Regarding Zcash specifically, I don’t think the argument should be “Zcash needs SHPH because Tor doesn’t work.” Tor is useful and Zcash already benefits from it.

The argument is that Zcash can benefit from having another transport option for situations where an operator needs something they can self-host and control. For example, a wallet or node operator could use SHPH as an additional connectivity path when conventional Zcash connectivity is being interfered with; infrastructure operators could use it for protected node/RPC connectivity; and deployments in hostile networks could operate their own SHPH endpoints rather than relying entirely on publicly known infrastructure.

And to be clear, SHPH isn’t claiming to provide Zcash’s anonymity properties itself. It is a transport/security layer that can sit underneath an application or infrastructure deployment.

I also agree that the project should prove this rather than simply claim it. That’s why I think the most useful outcome of the proposed grant would be independently reproducible testing and a concrete Zcash integration prototype, rather than asking anyone to take the project’s claims on faith.

I appreciate the pushback. This is actually helping me identify where the proposal needs to be more explicit about the distinction between what SHPH already implements, what is being funded, and what remains experimental.

1 Like
10

I feel like “me” is Claude, because the post seems to be Claude trying to communicate with YOU, the user.

It’s literally telling you what your argument should be, because I did not mention that.

Further:

Literally who asked…

Oh my holy Jesus, WHEN did anyone compare SHPH’s anonymity properties to Zcash? That doesn’t even make sense! One is at the application layer, the other is at the transport layer!

Agree with WHAT? WHO? Is this a venture in trying to prove that I’m schizophrenic or what? Like I know I’m sleep deprived, but it cannot be this bad…

11

Mate as I said sorry because of my English, I use ai to assist me on the documents and also these replies, you are right that it literally answered the questions that you nor me asked, You are right and have fair point, I’m going to be honest, I live in iran and layer 4s are the most popular ones here, you know why ? Cause when the Gov goes halal net like a few months ago, it actually works, SHPH is layer 3 that actually has PQC, it actually can’t be cracked ( although that’s a big statement and I genuinely want people to point it out so if there is anything to fix, I do it )

12

Also our biggest and main point to apply for zcash, is that we genuinely want more people, more platforms to use self hosted infrastructure, I think you already know about the whole observers ?

13

You are NOT using AI to “assist you in documents”. You are using AI to draft whatever this is, and the AI is doing most of the work.

Second, “SHPH is layer 3” You mean this is not TCP/IP, but an alternative to TCP/IP? You are proposing to ditch the system that is run by ISPs, meaning SHPH is fundamentally unable to organically gain nodes, since no ISP’s fiber lines and routers support it?

Bravo.

14

This is so nonsensical it is insane. Zcash is a user base, not a development agency. Even if it were, your proposal is already (a) incoherent, and (b) useless for Zcash users.

15

I to be honest disagree, I’m not asking Zcash to become a development agency. I’m proposing an open-source project because I think Zcash users and infrastructure can benefit from having more ways to operate their own connectivity under censorship

16

I think there’s a misunderstanding on the L3 point. I’m absolutely not proposing to replace TCP/IP or asking ISPs to support SHPH. SHPH runs over the existing Internet connectivity — the ISP still provides the underlying IP network. The point is to provide a different secure/censorship-resistant path between endpoints on top of that connectivity.

And on the AI point: yes, I use AI heavily for drafting and documentation, especially because English isn’t my strongest language. But the repository, architecture, implementation, testing and technical decisions are mine. I’m happy for people to judge those directly.

If I’ve explained the layering badly over here, that’s fair criticism and I’ll clarify it.

18

Then it’s not Layer 3!!! Layer 3 is the network layer! You’re still Layer 4!

19

That’s actually the part I disagree with most. SHPH isn’t just an idea or a proposal for something I might build someday. It already exists as an open-source implementation, with the protocol, cryptographic components, transport code, testing and development history publicly available.

The grant is for taking an existing project through further hardening, verification and Zcash-specific integration — not for funding a PowerPoint idea.

Whether that existing implementation provides enough additional value over Tor/Nym to justify Zcash funding is absolutely worth debating. But I don’t think it’s fair to describe the project itself as hypothetical.

20

No, the project as described by you as a “layer 3 defense” is hypothetical.

What exists is Layer 4, which is functionally the same purposes as Tor, but unaudited and vastly inferior, with a negligible anonymity set.