01
Read the blocks.
A bounded scanner reads new confirmed blocks on your chosen chains. Saved progress lets it continue after a restart.
HOW IT WORKS + TRUST
Check it yourself.
No mystery score. SENTR turns observable chain activity into a small, readable alert with evidence.
02
Compare the evidence.
Approval logs, transfer events and direct native sends are checked against your watch history and cached public threat lists.
03
Make it actionable.
Open your inbox or enable Web Push. Each alert includes a transaction link and a practical next step.
PROTECT, SHOWN
Two ways to keep
a mistake small.
Tap a scene or use its buttons. Each one plays what the tool does, and what it cannot do.
Reduced motion is on, so this shows the end state. The buttons still switch states.
Reduced motion is on, so this shows the end state. The buttons still switch states.
05 / IN VIEW
What gets an alert.
+
New permissions
Token approvals and operator access. Large allowances are flagged; a new spender means new to this watch, not proven malicious.
+
Funds moving out
Standard token transfer events and successful direct native sends from the watched address.
+
Lookalike dust
Incoming zero or dust transfers whose sender matches the first and last four hex characters of an earlier recipient observed by this watch.
+
Listed counterparties
Interactions with addresses in the cached ScamSniffer list. A list match is evidence to review, not a final verdict.
06 / OUT OF VIEW
A lookout has limits.
×
Before a transaction lands
SENTR is not a transaction firewall. It cannot stop a transfer, inspect unsigned backend instructions or judge what you intended to sign.
×
Every possible movement
Internal native transfers, nonstandard token events and off-chain signatures are outside watch coverage. Watching starts when saved. Protect can separately scan up to 2,000 older blocks per request, with a visible range.
×
Instant or complete coverage
Public RPCs can fail or lag. The scanner waits two blocks and checks up to 1,200 blocks per chain per tick. Push delivery depends on your browser and push service.
Evidence over reassurance.
Inspect the scanner, reproduce the local tests and read the exact limits.
07 / PUBLIC INTELLIGENCE
Know where the signal
comes from.
Source dates are visible. Old intelligence stays labelled. No list can know every threat.
ScamSniffer
Cached public phishing addresses. The free source has a seven-day publication delay, so new drainers may be absent.
Checking cache status…
Inspect the source ↗MetaMask
A phishing domain list, not a wallet address list. Transactions do not reveal the website you visited. Check a website here without storing it.
Checking cache status…
08 / YOUR DATA
Watch less of you.
More of the chain.
No account to create. No email to hand over. The browser keeps the private access token for each watch. Keep using the same browser to open that inbox.
We store the watched address, chain choices, optional push subscription, alert history and scan progress. Derived recipient history is needed to detect lookalikes. We do not log IP addresses in the application. Infrastructure providers may keep their own connection logs.
Unwatch removes that watch, subscription, history and inbox from the active store in one click. Previously delivered browser notifications cannot be recalled. Other people can independently watch the same public address.
Privacy you can inspect.
Choose 7, 30 or 90 days. The entire watch, inbox, recipient history and subscription expire automatically. No email, no analytics and no application IP logs. Burner and stealth secrets stay encrypted in this browser and are never sent to the API.
Open browser developer tools: Network shows only local assets, the watch API, your wallet RPC and optional push registration. Application storage shows encrypted vault ciphertext. Inspect the source links below for expiry and deletion. No private key goes to the server. Infrastructure may keep its own connection logs.