Before you act · September 2026 · about a five-minute read
SaveSaveSaveSave
A clear risk check before you click, sign, send, or trust.
1. The problem
Crypto scams and AI-assisted scams often look ordinary until it is too late.
A token may be impossible to sell.
A wallet may be connected to a suspicious address.
A message may contain a phishing link.
A prompt copied from the internet may contain hidden instructions.
A person may sign or share something without understanding the consequences.
Almost every one of those has a document in front of it — a message, a post, a DM. By the time you have an address to check, you have usually already been persuaded.
SaveSaveSaveSave helps you pause and inspect the risk before acting.
2. What it checks
- Crypto addresses and supported tokens.
- Wallet-related risk signals.
- Pasted scam messages.
- Prompts copied from social media, Reddit, blogs, Discord, Telegram and other sources.
A prompt is text you give to an AI assistant. A risk signal is something we found that is worth your attention — not a verdict on its own.
Every scan returns one of four results, and explains what was found and what was not checked.
3. How to read the result
No covered high-risk finding was detected.
Something deserves a closer look before you continue.
A serious risk was detected. Do not proceed without resolving it.
The system could not complete enough checks to give a reliable answer.
A PASS is not a guarantee of safety. It means that nothing covered by the completed checks was found.
Insufficient data is a result in its own right, not a soft pass. If an address in your message has not been checked, the message cannot come back clean — the row says so instead of leaving a silent gap.
When a scan covers several things at once, the worst part decides the guidance. Two clean addresses never make up for one failing address. Nothing is averaged.
4. What the four saves mean
Save your funds
Spot important risks before you send or sign.
Save your time
Run several checks in one place.
Save your trust
See the evidence and the source behind a finding.
Save the regret
Pause before a mistake becomes expensive.
5. Messages and prompts
You can paste a message or prompt copied from the internet.
The scanner looks for:
- Phishing and impersonation.
- Requests for seed phrases, private keys, passwords or codes.
- Suspicious links.
- Requests to connect a wallet or sign something.
- Hidden instructions.
- Attempts to manipulate an AI system.
- Obfuscated or invisible text — characters that take up no space on screen but are still read by software.
- Crypto addresses contained in the message.
Any address it finds is shown with the sentence it came from, so you can see how it was being used. You choose whether to check it.
The message scan is performed in your browser. If you choose to check a crypto address found inside the message, that address is sent to the relevant security provider for address analysis. The full message is not sent for that address check.
No scan can guarantee that a message or prompt is harmless.
6. Crypto coverage
| Network | What is checked |
|---|---|
| EVM chains | Token and wallet risk signals where supported |
| Solana | Native token security signals and wallet checks |
| Sui | Token security checks |
| TRON | Token security and address checks using the supported provider data |
| Other networks | Not currently covered |
Bitcoin is not covered. Neither are Cosmos-family chains or Cardano. If you paste a Bitcoin address, the scanner will not offer to check it rather than guessing.
Chain-by-chain detail is in the Technical Transparency document.
7. What the product does not do
- It is not financial advice.
- It does not predict prices.
- It does not guarantee safety.
- It does not prove that an address owner is criminal.
- It does not replace your own judgment.
- It does not ask for your private key or seed phrase.
- It does not execute prompts or messages.
- It does not automatically visit links found in submitted content.
- It does not currently perform a full JavaScript security audit.
- It does not currently run uploaded files in a sandbox.
8. Privacy
SaveSaveSaveSave is designed to minimise the data it receives.
Message and prompt scan
- Analysed locally in your browser.
- No full message or prompt is sent to the address-security provider.
- No seed phrase or private key should ever be submitted. Nothing here needs one.
Address scan
- The address must be sent to the relevant security provider to obtain external risk data. There is no way to look up external data without sending the thing being looked up.
- The provider and the scope of that request are named in the Technical Transparency document.
- We do not claim that third-party processing is equivalent to local-only analysis. It is not.
The submitted content scan is local. The website itself may still use limited operational services that do not receive the submitted scan content.
Today that means two things: a cookieless page-view counter, and Google's advertising code. Both see that a page was opened. Neither receives anything you paste, scan, or find — see section 9.
9. How this is paid for
The site costs about a dollar a month to run. It is not profitable and it is not self-sustaining.
This site carries advertising. Google AdSense, in two side columns on wide screens. They are labelled, they do not appear on phones, and they sit outside the page content rather than inside it.
We monetise the attention around the tool, not the information you bring to it.
Nothing you paste, scan, or find is ever sent to an advertising system. Not the text, not a finding, not a verdict, not an address. The ad code is given the page address and nothing else, and an automated test fails the build if that stops being true.
Advertising means the page loads code from Google, and Google can see that a browser opened a page on this site — the same as on most of the web. That is a real cost and we would rather state it than bury it. It does not change what happens to your scan, which is still analysed in your browser and still never transmitted.
Sponsors, if there are ever any, will not influence a verdict, and we will not take sponsorship from a project this scanner would reasonably need to assess.
10. The promise
Most tools ask users to trust a score. SaveSaveSaveSave tries to show the evidence before the decision.
Pause first. Check the evidence. Then decide.
Everything technical — the chains, the data providers, the sanctions oracle, the test suite, the known gaps and the exact privacy verification — is in the Technical Transparency document. You do not need to read it to use the product. It exists so that anyone who wants to check our claims can.