3 ms·

Still I don't previously remember Claude constantly trying to stop conversations or work, as in "something is too much to do", "that's enough for this session,

Still I don't previously remember Claude constantly trying to stop conversations or work, as in "something is too much to do", "that's enough for this session, let's leave rest to tomorrow", "goodbye", etc. It's almost impossible to get it do refactoring or anything like that, it's always "too massive", etc.

egeozcan 6mo ago
Even superpowers started dividing things into "phases". "I think we can postpone this to phase 2 and start with the basics". Meanwhile using more tokens to make a silly plan to divide tasks among those phases, complicated analysis of dependency chains, deliverables, all that jazz. All unprompted.
colordrops 6mo ago
I thought I was tripping when I saw this. Must have been a measure to reduce usage to save them some compute.
OccamsMirror 6mo ago
Not to mention the amount of placeholders and TODOs it's leaving in the codebase but then declaring that it's finished the work. I've cancelled my subscriptions to both Codex and Claude and am going to go back to writing my own code. When the merry-go-round of cheap high quality inference truly ends, I don't want to be caught out.
darkteflon 6mo ago
I keep reading about this, but I have never, ever seen it. Daily Claude Max user for ~6 months. Not saying it doesn’t happen, but it’s never once happened to me.
- pas 6mo ago
  A really big unknown in all of these anecdotes is what skills people have installed (and what's in their CLAUDE.md, and ...)
  - JohnMakin 6mo ago
    This really is a lot of it, at least trying to help people at work internally. I've discovered a lot of people overly rely on Claude writing directives (always do X, never do Y, remember this every time) to its MEMORY.md, which it does mostly unprompted. The problem is, the few times I've noticed my agent getting "squirrely," some or a lot of the stuff in MEMORY.md was flat out wrong (the agent wrote down the wrong memory), confused or in direct contradiction with its CLAUDE.md, etc. When I fixed this, it was like magic, working how I wanted again. I now have a skill to periodically audit MEMORY.md and CLAUDE.md according to the conventions I've learned work best for me - which I suppose /dream is supposed to handle eventually, but you're kind of trusting it to audit its own memories, which have, at least to me, already proven to be unreliable. With so many factors like this, not even to mention context exhaustion, window size, effort, etc. - anecdotal evidence is almost worthless without examining someone's entire local state. A lot of it, to me, feels like user error, I haven't really noticed much behavioral difference between 4.5, 4.6, and 4.7, at least in my own workflow. I will note though that constantly managing these things is a lot of work that I hope one day becomes less necessary. It's more than I can expect people on my team to manage on their own, and unless I sit down with them 1 on 1 and review their issues, or write some clever agent to help them, I don't really know how I can help people reporting things that I hear posted here a lot.
- codethief 6mo ago
  Hasn't happened to me, either, but this GitHub comment is reporting similar observations: https://github.com/anthropics/claude-code/issues/42796#issuecomment-4194012440 https://github.com/anthropics/claude-code/issues/42796#issue...

egeozcan 6mo ago

Even superpowers started dividing things into "phases". "I think we can postpone this to phase 2 and start with the basics". Meanwhile using more tokens to make a silly plan to divide tasks among those phases, complicated analysis of dependency chains, deliverables, all that jazz. All unprompted.

colordrops 6mo ago

I thought I was tripping when I saw this. Must have been a measure to reduce usage to save them some compute.

OccamsMirror 6mo ago

Not to mention the amount of placeholders and TODOs it's leaving in the codebase but then declaring that it's finished the work. I've cancelled my subscriptions to both Codex and Claude and am going to go back to writing my own code. When the merry-go-round of cheap high quality inference truly ends, I don't want to be caught out.

darkteflon 6mo ago

I keep reading about this, but I have never, ever seen it. Daily Claude Max user for ~6 months. Not saying it doesn’t happen, but it’s never once happened to me.

pas 6mo ago
A really big unknown in all of these anecdotes is what skills people have installed (and what's in their CLAUDE.md, and ...)
- JohnMakin 6mo ago
  This really is a lot of it, at least trying to help people at work internally. I've discovered a lot of people overly rely on Claude writing directives (always do X, never do Y, remember this every time) to its MEMORY.md, which it does mostly unprompted. The problem is, the few times I've noticed my agent getting "squirrely," some or a lot of the stuff in MEMORY.md was flat out wrong (the agent wrote down the wrong memory), confused or in direct contradiction with its CLAUDE.md, etc. When I fixed this, it was like magic, working how I wanted again. I now have a skill to periodically audit MEMORY.md and CLAUDE.md according to the conventions I've learned work best for me - which I suppose /dream is supposed to handle eventually, but you're kind of trusting it to audit its own memories, which have, at least to me, already proven to be unreliable. With so many factors like this, not even to mention context exhaustion, window size, effort, etc. - anecdotal evidence is almost worthless without examining someone's entire local state. A lot of it, to me, feels like user error, I haven't really noticed much behavioral difference between 4.5, 4.6, and 4.7, at least in my own workflow. I will note though that constantly managing these things is a lot of work that I hope one day becomes less necessary. It's more than I can expect people on my team to manage on their own, and unless I sit down with them 1 on 1 and review their issues, or write some clever agent to help them, I don't really know how I can help people reporting things that I hear posted here a lot.
codethief 6mo ago
Hasn't happened to me, either, but this GitHub comment is reporting similar observations: https://github.com/anthropics/claude-code/issues/42796#issuecomment-4194012440 https://github.com/anthropics/claude-code/issues/42796#issue...