A very interesting news about Bitcoin just came out — Robin Linus (, the guy from ZeroSync and BitVM), launched something called Binohash that could significantly change the game for smart contracts on the network. Basically, it’s a way to make Bitcoin Script "read" transaction properties without needing a soft fork or changing consensus rules. It sounds like science fiction, but it’s very real.

The problem that no one could solve was simple: Bitcoin Script is deliberately limited. It cannot directly access details like inputs, outputs, amounts, or other transaction fields. This makes it very complicated when building advanced protocols — like in BitVM bridges between Bitcoin and other chains. You need to prove that a specific transaction occurred with certain properties (a peg-out to the correct address, for example), and without that, you rely on trusted oracles or lightweight clients, which involve assumptions of honesty.

This is where Binohash comes in. The solution is quite creative: it creates a collision-resistant hash of the transaction that can be calculated and read directly within the Script. Because it’s collision-resistant, it’s impossible to fool the system by swapping one transaction for another with the same hash. It’s like a "transaction ID" document.

How does it work in practice? Robin explored two old behaviors in Bitcoin’s legacy opcodes — specifically OP_CHECKMULTISIG with its FindAndDelete step. Basically, when verifying signatures, the system removes all provided signatures from the scriptCode before calculating the sighash. Robin cleverly uses this: he places many "fictitious" pre-defined signatures in the lock script, and the spending transaction selects a subset of them. Different subsets produce different scriptCodes, which generate different sighashes. The spender then "grinds" (test many combinations) until the sighash satisfies a proof-of-work puzzle.

The numbers are impressive. With parameters W₁ = W₂ = 42 bits of work, you get about 84 bits of collision resistance — very strong for most use cases. An honest user only needs about 44.6 bits of grinding, costing less than US$50 in cloud GPUs. A real transaction has already been mined on Bitcoin’s mainnet demonstrating exactly this.

The implications are huge. This enables trustless introspection for BitVM bridges — you can verify peg-ins and peg-outs, state differences, all without oracles or full light clients. It’s basically bringing covenant-like functionalities to Bitcoin without changing consensus rules. It’s the kind of innovation that shows why Bitcoin continues to evolve in ways no one expected.