Binohash: Transaction Introspection Without Softforks A first transaction has been mined demonstrating a limited form of covenants using only existing Bitcoin script functions. This development potentially eliminates the need for a Bitcoin light client in BitVM bridges, simplifying the construction and improving security. We hope this development furthers research on the possibility of creating introspection and covenants-style functionality on Bitcoin without the need for softforks, and advances discussions on the usability of covenants in Bitcoin and the use-cases they enable. Many thanks to everyone at @idealgroup, as well as @PortlandHODL and the Slipstream team at @MARA for patching their nodes to mine these wacky transactions. Full paper in the comments 👇

BitVM2 made Bitcoin bridges possible. BitVM3 cut on-chain costs by 1,000x with garbled circuits. Now @liameagen's breakthrough scheme, Argo, cuts off-chain costs by another 1,000x—making off-chain computation on Bitcoin truly efficient and practical. 👷🏻‍♂️👷🏻‍♂️👷🏻‍♂️

The future of blockchain is Braids.

DMND Pool Now Open To All Miners, With SOC 2 Compliance and Stratum V2 Support bitcoinmagazine.com/bitcoin-… via @bitcoinmagazine

Congratulations to Braidpool contributor and @bitshala_org grantee @zaidmstrr on getting his second PR merged into Bitcoin Core! x.com/KevKevPal/status/19909…

✨New follow-up bitcoin-core change opprotunity✨ 📢mentioned in: #32821 rpc: Handle -named argument parsing where '=' character is used github.com/bitcoin/bitcoin/p… "> I'm wondering if there's a different way we can go about this without adding another table of arguments that need special treatment. I'm also not a fan of separate tables and suggested the following change to unify them earlier: b998cc52d51b48db9271fdba0bd69e9aaccb7999 ([tag](github.com/ryanofsky/bitcoin…)). This change is just a refactoring and could be a followup. > Perhaps we could move named argument handling and string to json conversion server side? I think moving logic server side would avoid need for duplicate tables, but not actually make the code or logic simpler because the current syntax for distinguishing named parameters is inherently ambiguous. (It probably would have been better to require named parameters to begin with `-` to avoid ambiguity. It could also be better to try to parse *every* argument as JSON and just fall back to passing strings if they are not valid JSON to avoid the need for the conversion table.) I feel like current PR just makes some small tweaks to parsing to make the current syntax work a little better, and it adds good test coverage. If we follow up this PR with b998cc52d51b48db9271fdba0bd69e9aaccb7999 or incorporate those changes here, the client code should be better documented and more maintainable too." - ryanofsky

My expectations of coding LLMs is growing far faster than their capabilities. I can run gpt-4o equivalents locally now with a dozen open source models. That's great. I no longer want that. What should I do? Suggestions encouraged.

Here is a statement, published on bitcoincore.org, signed by 31 contributors to Bitcoin Core, on their views on Knots vs Core. bitcoincore.org/en/2025/06/0…