Read 1 file, listed 1 directory (ctrl+o to expand)
Bash(echo "checking output works")
⎿ checking output works
Read 1 file (ctrl+o to expand)
⎿ API Error: 400 messages.3.content.56: `thinking`
or `redacted_thinking` blocks in the latest
assistant message cannot be modified. These
blocks must remain as they were in the original
response.
Very inspiring improvements. DIssapointing result for a code review i expected to see after my 30 min walk
I made the Slack onboarding step mandatory for now since we thought that a lot of our value was in sending investigations and PRs, and Slack was what we used ourselves.
What tool do you use for communication around your project? If you don't want to share publicly, could you please shoot a line to:
ash [at] superlog.sh?
Would love to learn about your usecase in more detail too!
Got it! What channel would you prefer instead? Would Telegram/WhatsApp/Signal/iMessage be good?
The platform itself doesn't need Slack to function, we just observed that users got more value if they could get notifications somehow, so I'm more than happy to add more comms platforms :)
I've started using ; in opencode because it seems more natural to string a couple of commands together without some awkward grammatical thing. Similar to the caveman prompt.
Em-dashes are too wide! I frequently forget what I've just read before I reach its other end. "--" in ASCII or en-dash + whitespace are superior separator symbols.
I am a believer that everyone should have their main flow be model/provider agnostic at a high level. I often run out of claude tokens and use GLM-5 as backup.
I tried the agnostic thing for a while, but there are enough quirks between the providers that I gave up trying to normalize it. GPT5.x wipes the floor with other models for my specific tool calling scenarios. I am not going to waste time trying to bridge arbitrary and evolving gaps between providers.
I put my Amex details into OAI, I get tokens, it just works. I really don't understand what the hell is going on with Claude. The $200/m thing is so confusing to me. I'd rather just go buy however many tokens I plan to use. $200 worth of OAI tokens would go a really long way for me (much longer than a month), but perhaps I am holding it wrong.
Being model and provider agnostic are orthogonal concerns.
e.g. you can run Claude models on AWS Bedrock giving you provider choice for the same model. Whether or not you need model agnosticism at that point seems like a very different question.
> . you can run Claude models on AWS Bedrock giving you provider choice for the same model
Is anyone doing this for personal dev that isnt token fed by employers? Coding plans are subzidized for a reason right? If I did API usage from a cloud provider id be out tens of thousands already.
There are differences for sure. Claude models feel the most 'stable' in that I see less tool confusion messages and other mistakes like the one im looking at right now.
"Wait, I'm editing the wrong sections. The edit tool tried to match but replaced with different prop names than what was in the file. Let me re-read the file and understand the current state properly."
And of course models are not 1-to-1 and have different strengths and weaknesses. I know I wont get the same quality plan mode output probably. Its a tradeoff.
I generally assume the differences could be minimized by tailoring the instructions to the models; that they're not incapable of doing the same things, but the way in which they're instructed matters because it needs to draw on training.
But I don't use any of the cloud stuff; I'm local4lyfe.
Around the part where Margaret explains the problem to Tom , and started to feel annoyed. I could tell it was a LLM trying to fit a sci fi novella style of writing. And it was doing a good job , it was certainly better than 90% of posts ive read in the last 6 months.
Dont know why that makes me annoyed, maybe cause its the depressing seriousness of being a 'prompter' and the americana framing of it.
Ill be honest i got all the way in the installer until it asked me to install Java 17+ manually. I havent installed java in forever and i remember it being a huge pain so ill pass. Im heavy on code orchestration tools atm and this one looks useful so hopefully when its further along in dev ill try it.
Hey, thank you for trying anyway. To be honest, I also remember installing java to be painful, but apparently nowadays brew install openjdk@21 + adding path to shellrc is doing the trick (in case you want to give it a second try)
You have to understand though that this is X (rip twitter) we are talking about and from the verified account, the 14k follower count, it is evident that this person either is or is trying to be a tech "influencer". Posting controversial rage-baits is pretty much the pattern every influencer follows today to stir up discussion, increase their visibility, and get more followers.
> I don't think he cares about being polite.
If you're polite, debate civilly, say reasonable things and act like a normal person, you are a nobody on X. Nobody will see your tweet. Nobody will engage with it. You might as well have not said anything.
Great work, my only nitpick is that when you quit, it tells you that its not gonna save, then its yes to confirm quit. I instinctively wanna type N to quit without saving.
I feel like thats the opposite behavior. For example, I use https://github.com/zyedidia/micro as my main editor in terminal, where it says `Save changes to filename before closing? (y,n,esc)”` when you try to quit.
Bash(echo test123) ⎿ test123
Very inspiring improvements. DIssapointing result for a code review i expected to see after my 30 min walkreply