(Follow-up to the original λPrize proposal.)
Quick update on Nabz, plus the part I owe this forum: an honest argument for why this belongs on the Logos stack rather than anywhere else.
The code is public, and it’s a working MVP
The full monorepo is now open source: GitHub - ibeezhan/nabz: Censorship-resistant daily alive-ping with Telegram, HTTPS, and DNS fallback transports. Identity via Telegram OAuth, X OAuth, or Ethereum wallet. · GitHub
It’s a runnable pnpm + Go workspace, not a slide deck:
-
packages/bot: Telegram bot (/start, /ping, /check). Transport layer 1.
-
packages/server: Fastify + Postgres core API for identity, ping ingest, signature verification, privacy, and status. Layer 2, plus the ingest sink for layers 3 and 4.
-
packages/dns-server: Go / miekg/dns authoritative resolver that decodes a ping from the query name itself. Layer 3.
-
packages/waku-transport: @waku/sdk light node, Light Push send + Filter subscribe over content topic /nabz/1/ping/proto, protobuf wire format. Layer 4, the Logos-native path.
-
packages/web: Next.js status dashboard (green / amber / red).
Tests pass (vitest across the four Node packages, go test for the DNS server), and CI runs the whole matrix plus a docker compose build on every push. The README has screenshots and a diagram of the full ping lifecycle if you want the 60-second version.
What Nabz actually is (and what it deliberately isn’t)
Nabz carries one bit: “this person was alive at time T.” No location, no message body, no metadata beyond what proves liveness. That narrowness is the whole design.
The problem it targets is specific and, unfortunately, recurring:
-
Iran, Nov 2019,2022,2025,2026. A near-total blackout during the protests. Families abroad refreshed dead apps with no way to know if relatives were alive, and rolling shutdowns have continued since.
-
Sudan, 2019 & 2023. Nationwide, weeks-long cutoffs imposed around mass violence, severing the diaspora exactly when contact mattered most.
-
Myanmar, post-2021. Some of the longest-running rolling shutdowns in the world, with whole townships dark for months.
Every existing tool misses the same way. Messengers, calls, and social posts assume a working, unfiltered path to a specific service, which is the first thing a censor drops. Mesh and offline apps assume the two people are physically near each other, but a shutdown usually separates someone inside the country from someone outside it. Satellite kit assumes hardware most people don’t have.
Nabz’s bet is narrower and more achievable: don’t carry a conversation, carry one bit, over whatever channel the censor forgot to close. It’s a harm-reduction tool for activists, journalists, and ordinary people under internet shutdowns. It is not a panic button, and not a substitute for real emergency infrastructure. I want to be precise about that so no one over-trusts it.
Transport-agnostic by design: a life-signal, not an app
The core idea worth stressing to this community is that the signal is transport-agnostic. A ping is a tiny signed assertion (user_id | date | timestamp | nonce, signed with ed25519 or an Ethereum EIP-191 wallet, plus a fresh anti-replay nonce). It doesn’t care how it travels. Today the client tries, in order:
-
Telegram (often whitelisted, proxied, or MTProto-tunneled)
-
HTTPS to any mirror
-
DNS, where the ping is the query name ([sig].[nonce].[uid].[date].p.nabz.TLD). Breaking DNS breaks the censor’s own infrastructure, so it’s usually the last thing standing.
-
Waku Light Push, with no trusted server, IP, or domain anywhere in the path.
The client only needs one to escape. And because the payload is transport-independent, the fallback list is meant to keep growing toward the genuinely off-grid:
-
Waku as the primary censorship-resistant path (already implemented)
-
LoRa long-range radio for the hop out of a dead zone to a node that still has a route
-
Meshtastic off-grid mesh radio, so a ping can hop device to device across a blackout region until it reaches connectivity
-
more resilient transports as they prove out
The point is graceful degradation. As each centralized route is cut, the same one-bit signal falls through to a less convenient but harder-to-block carrier, all the way down to radio that needs no internet at all.
Why the Logos basecamp is the right home for this
I’ve looked hard at where this should live, and Logos is the only stack whose design goals are the same as Nabz’s threat model rather than adjacent to it.
Waku is already load-bearing, not decorative. Layer 4 is the only layer with zero centralized dependency in its data path. It runs a Waku light node, uses Light Push to publish without operating a relay, and any observer uses Filter to subscribe to /nabz/1/ping/proto. Both are cheap enough to run on a phone or a tiny VPS. Waku is the most battle-tested piece of the stack (it’s what Status runs on), which is exactly what you want under a “libp2p peers are reachable but everything else is blocked” scenario. RLN-style rate-limiting is a clean answer to the obvious “won’t a public ping topic get spammed?” concern.
Nomos (the Logos Blockchain) closes the last trust gap. Right now the public-key to identity binding still lives in the Nabz database, a residual centralized dependency I’m not happy about. The roadmap item is a Nomos on-chain registry so identity registration and ping verification are anchored on-chain, and a ping can be verified with no trusted Nabz server in the loop at all. Nomos’s censorship-resistance and privacy-preserving posture (Cryptarchia PoS, the Blend mixnet for metadata privacy) line up directly with a threat model where the metadata of who-pinged-whom is itself sensitive. I’ll note honestly that these privacy components are early. The Logos testnet v0.1 (Mar 2026) is an integration and validation milestone, and several of these pieces are explicitly still in development, so I’m treating the Nomos registry as a roadmap target, not something to claim as done.
Codex (Logos Storage) is the natural home for durable, censorship-resistant public ping history: an append-only, operator-independent record of liveness that outlives any single server, retrievable by CID.
And the packaging fits. Logos Basecamp, a self-hostable distribution you run on your own hardware, is philosophically identical to how Nabz is meant to be deployed: resilience comes from there being many of these, run by people you trust. A Nabz observer or relay is a natural Basecamp module: subscribe to the ping topic, verify, expose status. Self-sovereign, private by default, transport-diverse. That’s not a marketing overlap, it’s the same architecture.
In short: Nabz is a small, sharp application of exactly what Logos is building the substrate for. It uses Waku today, it wants Nomos and Codex next, and it degrades all the way down to radio, which is the whole point of a proof-of-life signal that has to survive the worst day of someone’s life.
Happy to take feedback on the Waku integration, the anti-replay and verification model, or where a Nomos registry would slot in. The repo is open, and issues and PRs are welcome.