Why use SiriScore¶
Every Bitcoin transaction is public forever. Before you broadcast one, SiriScore tells you what it reveals about you — so you can fix leaks while they can still be fixed.
Why check before broadcasting¶
Once a transaction is on-chain it cannot be edited or recalled. Chain-analysis firms and curious observers apply well-known heuristics to cluster addresses, identify change outputs, and link transactions back to a single wallet. Most privacy leaks are cheap to avoid before signing — reorder outputs, avoid a round amount, skip a reused address — and impossible to undo afterwards.
SiriScore automates that pre-broadcast check: it runs the same style of heuristics an adversary would, and reports them to you first.
What a transaction can leak¶
A few examples of what observers infer from a single transaction:
- Which output is your change. If one output uses the same script type as the inputs and the other doesn't, or one amount is suspiciously round, the change output — and therefore your remaining balance — is exposed.
- That your addresses belong to one wallet. Spending many inputs together, or paying to an address that has been used before, lets observers merge address clusters.
- Which software built the transaction. Input/output ordering, locktime behaviour, and dust handling fingerprint specific wallets.
Each leak SiriScore detects maps to a documented heuristic with a weight and a suggested fix.
Who is it for¶
- Bitcoin users who want to understand whether a transaction they are about to broadcast leaks information about their wallet
- Wallet developers who want to surface privacy warnings before the user signs
- BTCPay Server operators who want to score transactions as part of a payment flow
- Privacy researchers who want a programmatic interface to Bitcoin transaction analysis
What it does not do¶
- It does not broadcast transactions
- It does not store or transmit your transactions to any third party by default
- It does not replace a full chain-analysis tool; it focuses on pre-broadcast, single-transaction analysis
Next steps¶
- Use the web app: try it with no install
- Install the library: Python API and CLI
- Run locally: keep everything on your machine