Roadmap

Four phases. One graph. Real users.

Anti-sybil infrastructure for Telegram Mini Apps and TON, shipped in phases. We publish what's done, what's in flight, and what comes next — without making promises we can't keep on dates.

Beta pilot is opening — applications via the form on the home page

Phase 1

Live

Foundation

Public site, brand, customer discovery. Validating the problem with real teams.

Marketing site live · guard.nexalix.io
Customer interviews with TON Mini App teams
API skeleton — Telegram initData verification, API-key auth

Phase 2

In progress

Beta pilot

First 3–5 teams onboarded for free in exchange for usage logs and feedback. The cross-merchant signal graph starts here.

Telegram + TON + Behavior signal collectors
Rule-based scoring with sub-200ms response
POST /api/v1/score, POST /api/v1/feedback
Pilot dashboard — keys, history, stats
npm SDK package for client integration

Phase 3

Up next

Cross-merchant graph

Sybils flagged in one project automatically catch in others. The network effect that no single-team OSS release can replicate.

Cross-merchant blocklist surface in API responses
Custom rules layer for advanced merchants
Appeal flow for end-users wrongly flagged
Public benchmark — anatomy of TON bot farms

Phase 4

Planned

Public launch & ML

Pricing announced, model trained on labeled pilot data. Hybrid rules + GBM scoring becomes default.

Pricing — free tier, pay-per-call, enterprise
ML model trained on pilot labels
Hybrid scoring (rules + GBM) in production
Score breakdown with feature importances

How we ship

No fixed dates. Just gates.

Each phase has a gate. We don't ship the next phase until the current one's metrics close.
No promised dates here. Real-world signal sets the pace, not a calendar.
Customer feedback overrides our preferences. If the rules-based scoring is good enough, ML waits.
Cross-merchant graph is the moat. Privacy, retention, and aggregation policies are documented up front.

Want to be in the pilot?

We're picking the first 10 teams. Free during the pilot in exchange for usage logs and feedback.

Apply for pilot Or write to hello@nexalix.io