BTC Medusa scans every coin in your wallet against 31 open-source privacy heuristics, including address reuse, change leaks, exchange links and transaction entropy, then shows you exactly how exposed you are. The scan runs blind: we never learn which coins you hold, what you ask, or who you are. No node required.
Bitcoin is public and permanent. Every payment leaves structure behind, and surveillance firms have spent a decade learning to read it. They cluster your addresses, spot your change outputs, and tie your coins back to the exchange that touched them. Most people have no idea what their wallet is already saying about them out loud.
The single biggest privacy killer. Every reuse merges your activity into one identity.
Spending two coins together tells the world they belong to the same person.
Address-type mismatches and round amounts quietly reveal which output is your change.
Coins are matched against 364 known services and 30M+ labelled addresses.
Boltzmann analysis measures how many interpretations of a transaction actually exist.
Version flags, input ordering and signature quirks can identify the software you use.
BTC Medusa runs the same battle-tested heuristics behind am-i.exposed against your UTXOs and returns a plain 0–100 score, a letter grade, and a list of exactly what's leaking, and how to fix it.
Three of your receive addresses appear in more than one transaction, collapsing them into one cluster.
A change output traces two hops back to a deposit address at a major exchange.
The deterministic link between inputs and outputs is unambiguous. Entropy ≈ 0 bits.
A round-number output makes the payment-vs-change split obvious to any observer.
The open-source engine is brilliant, but to use it privately you had a hard choice: leak every address you look up to a third-party API, or run a full node and self-host the whole stack. Most people can't, or won't. BTC Medusa removes the choice.
We take the open-source heuristic data, encrypt it, and pack it into the block filters your wallet downloads. You get the full privacy analysis without ever broadcasting what you're looking at.
We don't ask you to trust our server. We designed the whole protocol around the strongest possible assumption: that the operator, us, is actively hostile, colluding, and trying to deanonymize you. Under that assumption, here's what an attacker is up against.
Full control of our own software, database and network, and it still can't see your coins or your queries.
ISPs and state actors watching the wire see Tor traffic: no IP, no payload, no link to you.
Even if we hand everything to an exchange or chain-analysis firm, there's nothing in our logs to hand over.
A growing anonymity set plus per-request Tor circuits leave only a guess that decays as the user base grows.
You want to ask one question: "how exposed is this coin?" without anyone, including us, seeing what you asked. So your wallet scrambles the question before sending it. We answer the scrambled version without ever being able to unscramble it. Your wallet then removes the scramble and reads the answer.
α = k · H(input), which looks like pure noise.k and reads the answer. We never saw the question.β = v · α, without ever learning your input.DLEQ proof that proves we used the real key and didn't cheat.f(x) without ever seeing x, and proves it did so honestly.
After the blinding, the zero-knowledge proofs and the Tor transport, our entire view of you reduces to this. Almost every meaningful fact is simply never knowable to us.
| About you | Can we see it? | Why not |
|---|---|---|
| Which coin you're scanning | No | blinded before it ever leaves your device |
| What the result says | No | unblinded only inside your wallet |
| How many coins you hold | No | tokens are spent without a counter we can read |
| Your IP address | No | Tor hidden service, traffic never exits the network |
| Your identity | No | no accounts, no email, no sign-up |
| Whether two scans came from you | No | each request is cryptographically unlinkable |
| That some valid scan happened | Yes | by design, it's all we need to keep the system running |
It's the first thing a careful Bitcoiner is thinking right now, and it's a fair question. Just days ago a soundness flaw in Zcash's Orchard shielded pool, hidden for nearly four years, could have allowed the undetectable creation of unlimited counterfeit ZEC, forcing an emergency hard fork. But notice what it put at risk: the money supply, not anyone's privacy. So it's worth being precise about what zero-knowledge proofs actually do here, because in BTC Medusa they are not what protects you.
Your query is blinded on your device before it leaves, and unblinded only after it returns. That blinding is plain, well-understood elliptic-curve math: k · H(x). A completely broken proof system could not reverse it. Tor hides your network identity on a separate layer. Your privacy never rests on a circuit being bug-free.
Our zero-knowledge proofs do one mundane job: prove a query was paid for and hasn't already been spent, so customers can't cheat the meter or replay tokens. It's a billing and anti-fraud tool. Its worst-case failure is an accounting headache for us.
Zcash put zero-knowledge proofs in charge of money. We put them in charge of billing. Your privacy is guarded by math that doesn't depend on them.
Every cryptographic primitive, every circuit constraint, every protocol flow is open and auditable. You don't have to take our word that we can't see your data. You can read exactly why we can't.
Our launch release runs as a plugin for Sparrow Wallet on desktop. We choose Sparrow since it's one of the most popular and robust desktop wallets around. However, it has not been endorsed by its creator. In the future, we hope to have all wallets, including Sparrow, bundle our plugin natively, since a percentage of every subscription will go straight to the open-source development team.
Send us a message below. We're happy to walk through the cryptography, the threat model, or wallet integration, and your note reaches both of us directly.