Back to docs
MethodologyPrivacy by designDecision support

Trust & methodology evidence boundaries.

Tri-Proof Guard helps campaign teams make defensible reward decisions. It combines wallet evidence, campaign policy and reviewer workflow while keeping human judgment and privacy boundaries explicit.

On-chain evidence first
Wallet age, transaction history, funding source, contract diversity and campaign actions are treated as evidence, not identity claims.
Cluster review
Shared funding and similar behavior create review groups so teams can inspect linked wallets together.
ScamGuard signal
ScamGuard adds suspicious URL, token mint and transaction intent checks, but it does not replace wallet risk scoring.
Privacy boundary
Reports avoid raw personal data. Proof packages are campaign-scoped and explainable without becoming a global identity record.
Reason code dictionary

Compact labels for explainable decisions.

Reason codes make exports, API responses and customer reports easier to audit. Each code maps to human-readable evidence in the wallet drawer.

KNOWN_ENTITY

Exchange, bridge, protocol, service or public infrastructure wallet.

SHARED_FUNDING

Multiple wallets appear connected by funding origin.

CLUSTER_LINKED

Wallet belongs to a suspicious behavioral or funding group.

LOW_HISTORY

Not enough reliable on-chain history for automatic approval.

CAMPAIGN_ONLY_ACTIVITY

Wallet behavior is concentrated around campaign actions.

REQUIRES_REVIEW

Human team decision should happen before distribution.

REWARD_EXCLUDED

Current policy excludes this wallet from the clean list.

False-positive handling
Tri-Proof reports keep gray-zone wallets separate from rejects, support team notes, and retain override history so campaign teams can correct conservative or aggressive decisions.
ScamGuard boundary
ScamGuard is a pre-sign risk screen for Solana links, token mints and transaction intent. It is not a guarantee, does not replace Sybil scoring, and should be used as one signal among several.

Operational safeguards

A risk label does not prove fraud or malicious intent.
Known exchange/service wallets are marked for review because they may not represent individual campaign participants.
Team overrides are retained separately from original Tri-Proof scores.
Gray-zone decisions should use evidence, notes and second-reviewer checks for high-value campaigns.