Also what has been happening a long time is that if you try to do any opencode development Anthropic models will start replacing the word opencode with claude intermittently.
Imagine how difficult tool calling gets, when your ~/projects/opencode path gets intermittently replaced to ~/projects/claude during the roundtrip to Anthropic API
They have been fighting back a while already, eroding trust in their models as a price.
I was even able to have an absurd conversation with Claude about it, quite kafkaesque
Hey, so I have been tinkering lately with locally running LLMs for programming purposes, on a Strix Halo machine with 128GB of RAM. I quickly realised that in order to get more speed out of the setup I need to save on stuff that ends up in model context. I built (with agents!) this simple bash command runner that swallows large output, passing only the relevant stuff to agent.
Think pytest only outputting OK if all tests pass, but in case of error getting the necessary error output only.
It is a cli, basic use for agents:
oo pytest
oo cargo test
etc…
Custom patterns configurable and there is also learn -flag to use Anthropic models to have LLM write the configuration for your custom command.
Nothing too complicated or special, but works in my LLM workflow saving on token use & speeding models!
Nice idea, but the labels are a bit too opinionated for me.
Literally all my code has been ”ghostwritten” for the past 18 months. Does not sound like something enterprise customers would like to hear and try to understand what it means.
So happy to see this featured here! Had been tinkering with bikes a long time before finding Sheldon’s site, but when I did I was dumbstruck by the amount of insight. And to top that, what a person he was. RIP
They are a small team and tool has gotten wildly popular. Which is not to say that slowing down and addressing quality and security issues would not be a bad idea.
I’ve been an active user of opencode for 7-8 months now, really like the tool, but beginning to get a feeling that the core team’s idea of keeping the core development to themselves is not going to scale any longer.
Yes, I get it. Personally dancing between commercial work (GH) and personal (no need for GH). Because of orchestration setup works for former end up using also for latter.
Maybe should try to give tickets a go. gh cli does add another HTTP layer and slows things down - feels silly to be paying for Cerebras if one is slowed down by other tooling.
What I do not understand about beads or alternatives such as this is: why not just use github issues? Or is it just the aversion of being tied to github?
My agent team does, easy to tell them so and give them access to suitable gh commands.
Imagine how difficult tool calling gets, when your ~/projects/opencode path gets intermittently replaced to ~/projects/claude during the roundtrip to Anthropic API
They have been fighting back a while already, eroding trust in their models as a price.
I was even able to have an absurd conversation with Claude about it, quite kafkaesque