NexusVeritas Protocol

NexusVeritas resolves operator identity, computes behavioral fingerprints, and produces deterministic risk scores for Solana tokens and deployers. The protocol has three layers — each one builds on the last.

Identity Behavior Risk

Operator Identity

RFC-01 · OI_v1

Defines how NexusVeritas represents an operator across wallets, lifecycle states, and funder lineage. This is the foundational layer — fingerprinting and risk scoring both depend on it.

operator_id
Deterministic canonical identifier, stable across wallet additions and lifecycle transitions.
wallet_set
Ordered, append-only set of bound wallets. Unbinding never deletes history.
funder_lineage
Immutable record of funding origin, inherited by all wallet bindings.
lifecycle_state
new → active → dormant → restricted → closed
Implementation status — Solana v1

RFC-01 is a multichain-ready target spec. Current Solana production uses the deployer's public key directly as the identity key (single-wallet), with cluster_id in CL-XXX format. Multi-wallet binding and explicit lifecycle states are planned for a future release.

Behavioral Fingerprint

RFC-02 · VEC_v2 · LIVE

Every operator is encoded into a 25-dimensional behavioral vector across five feature groups, normalized per group and adjusted for temporal decay. Live in production across 2.8M+ operators.

Funding Layer5 features · [0-4]
Activity Layer5 features · [5-9]
Launch Layer5 features · [10-14]
Behavioral Layer5 features · [15-19]
Structural Layer5 features · [20-24]
Validation

At n=1012, 90.3% of items' nearest neighbor is within their own archetype. Z-score normalized per group, linear temporal decay, coverage gating at 0.5.

Risk Engine

RFC-03 · RE_v1 · LIVE

Delta-based deterministic scoring. Same input always produces the same output — no ML inference, no runtime heuristics, no synthetic data influence.

CRITICAL
score ≥ 85
HIGH
score ≥ 60
MEDIUM
score ≥ 30
LOW
score < 30

Final score combines operator and token signals with RE_v1 weights:

operator_risk = operator_score × 0.6
token_risk     = token_score × 0.4
final_score    = operator_risk + token_risk
Hard refuse

Known scam addresses trigger an immediate hard-refuse, overriding all other signals — highest priority in the risk contract.

Scan Endpoint

Single endpoint returns full operator context for any Solana token address.

GET /api/v2/scan/solana/{token_address}
Authorization: Bearer <API_KEY>

Response Format

{
  "score": 87,
  "risk_class": "CRITICAL",
  "operator_risk": 52.2,
  "token_risk": 34.8,
  "cluster_id": "CL-003",
  "archetype": "WALLET_FACTORY",
  "reasons": [
    { "text": "factory deployment pattern", "delta": 35 },
    { "text": "burst launch behavior", "delta": 24 }
  ],
  "is_hard_refuse": false
}

Errors

400Invalid token address format
408RPC timeout — retry
500Internal scan failure