Calling all stackers!
Leave a comment below to let the SN community know what you're working on this week. It doesn't matter how big or small your project is, or how much progress you've made.
Just share what you're up to, and let the community know if you want any feedback or help.
I've mostly been finalizing embedded wallet stuff. I've also been using pstack's
create-verfication-skillto provide more context for verifying changes in the browser.That's an interesting method to get rid of the now-standard "couldn't verify x, y, z because <excuse>".
That's my hope. It's very token-heavy to build, but it should make review more accurate.
I only scanned the instruction but it looked like it was a bit monolithic in its approach - would be great if you could run that on a diff. I.e. what I generally do is:
tag..tag) and group changes by logical areaSometimes at the diff level this can be already insane in terms of tokens but whatevs - I wouldn't want to do it against a monorepo.
I haven't tried using it much yet, but it creates verification recipes for features which should get picked up by review runs is my sense.
I watched a workshop from the author poteto describing how important it is to give LLMs better verification criteria. They also build a lot of static analysis tools to constrain the bot's output to less buggy modes, e.g. forbidding any use of
useEffectin react.Yes! I have developed this nasty (for me) habit of writing a long list of "acceptance criteria" which both claude and codex stock system prompts seem to pick up on well. It slows me down a bit, especially because when I get some highbrow result it's because I forgot to spec something properly in that section, but generally its where the difference is made. That's also where I try to guide plan generation quality but the indirection in that is often making me not like the resulting plan and editing/tuning it.
Not looking forward to that PR, lol!
Well, the whole workshop was about learning how to make agents trustworthy ... so that you can run thousands of agents without having to review all the code they generate. lol.
I do like the part about making agents more trustworthy though!
I'm just going to say "yes".
This week it’s my turn to work on my backups: checking and changing passwords, verifying 2FA, and all the backups. I just need to tweak a few details and that’s it. A light week.
Um...learning a language (a real one, not programming) and brushing up on my server management skills.
Doing manual integration tests in a browser, sucks energy, but has to be done.
I know that feeling. Manual browser testing can be exhausting, but it is usually where the weird bugs finally show up. Automated tests can tell you that individual pieces work, but the browser is where timing issues, state problems, permissions, and real user flow start interacting with each other. The worst part is when everything passes in isolation and then one small click breaks the whole flow. 😄 What are you integrating?
This week I am continuing a few small Bitcoin related experiments. I am still testing my solo mining applications. My Bitcoin miner has not found a block, which is no surprise given the odds, while on the much smaller MicroBitcoin network I have already found several solo blocks. That contrast has been a great way to understand mining probability and network difficulty in practice. I also started looking into AuxPoW and Namecoin merge mining after a discussion on Nostr. While working on mining, I built another small tool that watches network activity and gives me a sound alert when a block is found. I would love feedback on things like explorer support, wallet monitoring, and reliable block detection. The funny part is that every project seems to create another project. 😄 I am working alone, so I am mostly trying to finish small pieces, learn from them, and connect them over time. If anyone here has experience with AuxPoW, Namecoin, explorer APIs, or block monitoring, I would be happy to hear from you.