@lifeofjer Couldn't adding a pre tool hook have prevented this? I find AI agents sometimes miss documented instructions but hooks have proven effective. Not sure what Cursor's implementation of hooks is but you can configure hooks directly on Claude Code
Lol. When the Claude token usage bug happened last week and I decide not to do anything with it while it was being fixed it was as if I was on an holiday 😁
"Using coding agents well is taking every inch of my 25 years of experience as a software engineer, and it is mentally exhausting.
I can fire up four agents in parallel and have them work on four different problems, and by 11am I am wiped out for the day.
There is a limit on human cognition. Even if you're not reviewing everything they're doing, how much you can hold in your head at one time. There's a sort of personal skill that we have to learn, which is finding our new limits. What is a responsible way for us to not burn out, and for us to use the time that we have?" @simonw
Another Banger. At this rate do I need to try to solve other problems or goals I have 🤣. If I wait for one more week it probably would get solved. At this point we need to know Claude team's roadmap
You can now enable Claude to use your computer to complete tasks.
It opens your apps, navigates your browser, fills in spreadsheets—anything you'd do sitting at your desk.
Research preview in Claude Cowork and Claude Code, macOS only.
Claude is the perfect complement for me:
Whenever I have a question I can't answer, I ask Claude, and it gives me the perfect answer every time.
But as soon as I ask Claude something I do know, the answer is usually horseshit.
We just released Claude Code channels, which allows you to control your Claude Code session through select MCPs, starting with Telegram and Discord.
Use this to message Claude Code directly from your phone.
Over time, this reduces the noise.
You stop reacting to every release
and start recognizing what actually matters within minutes.
So it looks like I’m keeping up with everything…
But really, I’m just filtering better.
From there, I operate on 3 layers:
Depth (one anchor)
Selective expansion (only tools that improve my workflow)
Fast pattern recognition (try → classify → move on)
At the start of 2026, I stopped trying to follow everything in AI.
Instead, I anchored my learning to one company: Anthropic.
Not because it’s “the best”, but because depth beats scattered awareness.
Once you understand one well,
you can transfer that mental model across others fast.
You’re no longer learning from scratch
you’re mapping differences.
Part of that decision:
Anthropic was leading in how they structured frameworks for building with AI in 2025
and they’ve continued to execute into 2026.
That made it a solid foundation to learn from.
@gavinpurcell I have been using this all day and aside from it being slow there isn't a feedback that shows its working...that may be the difference between the experience with Dispatch and Claude Code