Start here
The free tool below runs entirely in this browser — no account, nothing stored or transmitted. Redact anything you would not want kept before you type it.
Open the companion tool →Why the destination is the only part that cannot bluff
A text message can claim any sender. The displayed name is just characters the sender typed. The only part of the message that has to be technically true is where the link actually points — and that is the part most people never look at.
Toll agencies, utilities, and delivery companies all publish their own account or payment pages, reachable without the text. Checking there takes about a minute and defeats the entire message, whatever it says.
The protocol
- Do not tap the link. Look at the domain in the message instead — the part immediately before the first single slash.
- Compare it with the agency's published domain. Lookalikes usually change one word, add a hyphen, or use a different top-level ending.
- Open the agency's own site or app by typing the address yourself, or use the number printed on your bill or account statement.
- Check the account there. If a balance or fee is real, it will be visible on the account itself — not only in a text.
- Report the message to your carrier by forwarding it to 7726 (SPAM), then delete it.
What the tool does
Paste the text you received and get a reading of the claimed sender against the real link destination, plus the independent route to the agency's own account page. It runs in this browser; nothing is sent or stored.
Limit first: This is a message-reading aid, not a fraud investigation service. It can flag a legitimate notice and it can miss a cleverly built link. A clean reading never proves a message is genuine.
Open the companion tool →One human next step
If you already paid through a link, call your card issuer to dispute the charge and report the message at ReportFraud.ftc.gov. Keep the number and message; both help the next person's case.
Research log and safety checks
AI-assist path: not used for a verdict. The artifact is a domain-comparison reading the reader can audit line by line; an opaque model judgment would be strictly worse than showing the mismatch. §6.10: not a frontier-pace topic.
§0.5 protection result: category is scam-fraud-defense; crisis-adjacent: no; fear/urgency/scarcity toolkit used: no. The protective function is free and the paid feature is convenience only.