Filtering Spam Tokens in Wallet Interfaces
Design reversible spam-token filters using explicit signals, without treating a token symbol, unsolicited transfer or logo as proof of trust.
Overview
When translating these concepts into an integration, keep the scope described by the wallet history api with the result. The examples in wallet transaction history and smart contract event logs show how the same distinction affects neighbouring data questions.
Wallets can receive unsolicited tokens and events without their owners asking for them. A useful interface reduces noise while preserving the ability to inspect the underlying records and challenge a classification.
Use multiple signals and clear labels
An unknown token, suspicious metadata or an unsolicited transfer can be a review signal. None alone establishes malicious intent. Separate unverified, hidden and confirmed policy-blocked classifications. Never trust a symbol, logo or metadata link as an ownership credential. Asset identity must remain anchored to the network and contract rather than a familiar ticker.
Worked example: a familiar symbol
An unrelated contract uses the same symbol as a widely recognised token and sends a tiny balance to many wallets. The UI does not merge the holding with the recognised asset. It shows or hides the contract under its own identity, preserves the evidence and avoids automatically opening promotional links embedded in metadata.
Make filtering reversible
Track the reason, rule version and user override where appropriate. Test a legitimate newly issued token, a zero-value event and a small but real holding. Keep filtered assets out of headline portfolio totals only under a documented valuation policy. Display filtering should not silently delete export records or rewrite historical balances.
A practical acceptance test
Give two contracts the same symbol and logo text, but mark only one as a verified identity under the fixture policy. Their balances and prices must remain separate. Apply and reverse a user display override, then verify that raw exports and event evidence retain the original asset identities and do not inherit another contract’s trusted label.
Questions about the topic
Frequently asked questions
Does a matching symbol identify the same token?
No. Symbols are not globally unique asset identifiers.
Should hidden tokens disappear from raw exports?
Only under an explicit export policy; preserve access to the original records.
Standards & technical documentation
