141–150 of 172 posts
I can only recommend to regularly measure how many tokens a harness+model combination uses for a certain task There are huge token efficiency/bloat differences between agents while working on the same tasks, using the same model, in the same environment Yesterday I ran 10 agentic tasks using GPT 5.6 Sol in an ubuntu 26.04 vm a couple of times with different harnesses and got vastly different token usage. +-----------…
I was an early and passionate adopter and paying customer of Cursor (since 2023!), but it’s probably been 6 months since I opened it. These days I “write” code with claude code and codex, and read/review it on GitHub. If I need to read it locally, I use a plain text editor. Can someone help me understand what value cursor offers in 2026?
Earlier quoted context omitted.
Cursor has an import assistant from vs code settings. Ahere is no assistant for the inverse. Not even for moving between computers (at least a couple of months ago when I had the need haha).
But doesn’t vscode’s settings.json have most if not all settings stored inside it? Should be a simple copy and paste.
I was an early and passionate adopter and paying customer of Cursor (since 2023!), but it’s probably been 6 months since I opened it. These days I “write” code with claude code and codex, and read/review it on GitHub. If I need to read it locally, I use a plain text editor. Can someone help me understand what value cursor offers in 2026?
Cursor has two benefits 1) It's still an IDE. Because of pricing I mostly use Codex, but I always have a VSCode/Cursor IDE open, thus have to juggle between the two. Working directly in the IDE is more comfortable. For full on vibecoding that might be worse, but when you want to do a deep review of the changes, an IDE is way better than reading a diff on github. 2) It supports every model. It's often very helpful to…
They seem to be doing their darndest to kill advantage #1. Even the command line "cursor ~/git" no longer opens an IDE in the ~/git folder but some random Chatgpt-esque interface that nobody asked for. Honestly, VS Code is a better "old Cursor" now than Cursor is.
#2 hasn't been much of an issue for me since Opus 4.5 came out. Anything the other models could do, Claude could do better. Of course things are starting to change now, and I've noticed Sol Ultra is specifically better at visual design stuff than Claude, but even that ~3% of cases I switch to GPT are still more easily solved with a cheap Codex sub than Cursor.
I was an early and passionate adopter and paying customer of Cursor (since 2023!), but it’s probably been 6 months since I opened it. These days I “write” code with claude code and codex, and read/review it on GitHub. If I need to read it locally, I use a plain text editor. Can someone help me understand what value cursor offers in 2026?
Every 30 seconds in claude you get a new, elaborate, contrived bash command that claude wants to run. It's some combination of 10 things you've already granted permission for, but this one is new!
Cursor's auto-review run mode fixes that https://cursor.com/docs/agent/security/run-modes
I know it's a compromise in security, but auto-review+sandboxing provides a much nicer out-of-the-box experience. I'm happily using Cursor + Composer 2.5/Opus for most tasks.
I was an early and passionate adopter and paying customer of Cursor (since 2023!), but it’s probably been 6 months since I opened it. These days I “write” code with claude code and codex, and read/review it on GitHub. If I need to read it locally, I use a plain text editor. Can someone help me understand what value cursor offers in 2026?
I'm surprised that their biggest feature has not been mentioned: The permissions model is much less noisy in Cursor. Every 30 seconds in claude you get a new, elaborate, contrived bash command that claude wants to run. It's some combination of 10 things you've already granted permission for, but this one is new! Cursor's auto-review run mode fixes that https://cursor.com/docs/agent/security/run-modes I know it's a co…
Earlier quoted context omitted.
> Claude Code injects a ton of tools into the system prompt It's so unfortunate they don't let you use the subscription with other harnesses anymore - since even if I used OpenCode they'd still get a bunch of useful data from the API calls, meanwhile I could stretch their tier limits way further.
It's so unfortunate people don't realize it's cheaper to write their own well working agent instead of insisting with general purpose bloated ones like Claude. 200$/month is a lot of money on Luna/DS4 flash, like really a lot and the results are much better than clowning on bloated CC. It's absurd how you have more and more organizations encoding their processes on LLMs and "engineers" (charlatan coders) don't even b…
Earlier quoted context omitted.
in my book anything that is (I'm sure well intentioned) and injected to help the agent — but doesn't help it — is a waste of tokens but even injected context that when I read it sounds useful can oversteer the model and make it second guess or take a more complicated route than it normally would (you can see this when looking at traces with and without that injected context) often harnesses also mention in their syst…
I am super curious how much in your opinion the extra prompting is useful? >often harnesses also mention in their system prompt locations of markdown files that the model can consult if the model thinks they might help >that hint alone as part of the system prompt can be strong enough to make the model read in more tokens than would have been necessary I purposefully do this as I imagine it is useful. In my project I…
Works fine. My conclusion is treat it like a starting point and not a holy bible of everything the model needs to know.
I pair this with making sure task files are commented with headers like build scripts so context+instruction is consistent.
The happy path then is the LLM pulling in context where it needs it.