Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Claude Code Routines

code.claude.com

Post reply on HN
Print view View original

281–290 of 451 posts

Re: Claude Code Routines

#281

Post by SoMomentary »

qingcharles wrote: post #177
pacha3000 wrote: post #116

Earlier quoted context omitted.

I'm the first to be tired of everyone, for every model, that says "uuuh became dumber" because I didn't believe them ... until this week! Opus is struggling worse than Sonnet those last two weeks.

Is it? Or is it the task you're trying to do? Opus 4.6 has been staggeringly good for me this last week, both inside Claude Code and through Antigravity until I used up my quota.

I think some of this comes down to undeclared A/B testing. I've had the worst week of interactions I have ever had using Claude Code. The whole week whenever I have a session that isn't failing miserably I seem to get tapped for a session survey but on any that are out and out shitting the bed it never asks. It has felt a little surreal. I'd love to see a product wide stats graph for swearing, I would 100% believe that it is hitting an all time high but maybe I'm just a victim of a bad A/B round.

Re: Claude Code Routines

#282

Post by stavros »

hyraki wrote: post #277
stavros wrote: post #207

Earlier quoted context omitted.

So basically what happened: 1. Anthropic realized their models weren't enough of a moat. 2. They built tools so they could expand their moat. 3. People don't want to use their tools, they want their models, and use other, better tools. 4. Anthropic bans the use of better tools, taking advantage of their model superiority to try to lock people into subpar tools. "I don't have enough of a moat so I'll use my little moa…

What are other better tools? I would like to use them

OpenCode, Pi, whatever Anthropic doesn't let you use with their subscription because they want to lock you in to their stuff.

Re: Claude Code Routines

#283

Post by rogerrogerr »

taytus wrote: post #92
dgellow wrote: post #50

Earlier quoted context omitted.

Their growth over the past months has been more than insane. It’s completely expected they don’t have the compute. You don’t have infinite data centers around

Like or not, openai isn't having the same compute strain, meaning this was predictable.

Or, Anthropic has better models and is experiencing higher demand because of that.

Or, OpenAI was reckless in securing compute.

Re: Claude Code Routines

#284

Post by latentsea »

danudey wrote: post #257
rbalicki wrote: post #165

Earlier quoted context omitted.

You can lessen your dependence on the specific details of how /loop, code routines, etc. work by asking the LLM to do simpler tasks, and instead, having a proper workflow engine be in charge of the workflow aspects. For example, this demo ( https://github.com/barnum-circus/barnum/tree/master/demos/co... ) converts a folder of files from JS to TS. It's something an LLM could (probably) do a decent job of, but 1. not n…

One thing my team lead is working on is using Claude to 'generate' integration tests/add new tests to e2e runs. Straight up asking Claude to run the tests, or to generate a test, could result in potential inconsistencies between runs or between tests, between models, and so on, so instead he created a tool which defines a test, inputs and outputs and some details. Now we have a system where we have a directory full o…

I'm looking at implementing https://github.com/coleam00/Archon as a means to solve this. You can build arbitrary workflows custom to your codebase. Looks to bring a bit of much-needed determinism.

Re: Claude Code Routines

#285

Post by chillacy »

taspeotis wrote: post #212
comboy wrote: post #107

Unrelated, but Claude was performing so tragically last few days, maybe week(s), but days mostly, that I had to reluctantly switch. Reluctantly because I enjoy it. Even the most basic stuff, like most python scripts it has to rerun because of some syntax error. The new reality of coding took away one of the best things for me - that the computer always just does what it is told to do. If the results are wrong it mean…

https://marginlab.ai/trackers/claude-code/

If SWE Bench is public then Anthropic is at a minimum probably also looking at their SWE bench scores when making changes, I'd trust more a tracker which runs a private benchmark not known to Anthropic.

Re: Claude Code Routines

#287

Post by yohamta »

Claude and Open AI seems to be trying not to be 'Just a model', but this is intrinsically problematic because model can be degraded and prices only goes up once they lock-in customers. It is increasingly important for anyone who are responsible of managing 'AI workflows' to keep the sovereignty about how you use AI models. This is why I'm super excited in building the local-first workflow orchestration software called "Dagu", that allows us to own your harness on your own. It's not only more cost-effective, but outcome is better as well because you have 100% full control. I think it's only matter of time that people notice they need to own their workflow orchestration on their own not relying on Anthropic, OpenAI, or Google.

Re: Claude Code Routines

#288

Post by FuckButtons »

cortesoft wrote: post #61
unshavedyak wrote: post #39

Earlier quoted context omitted.

Wait we can't use claude -p around other tools? What is the point of the JSON SDK then? Anthropic is confusing here, ugh. edit : And specifically i'm making an IDE, and trying to get ClaudeCode into it. I frankly have no clue when Claude usage is simply part of an IDE and "okay" and when it becomes a third party harness..

I was pretty sure that claude -p would always be fine, but I looked at the TOS and it is a bit unclear. It says in the prohibited use section: > Except when you are accessing our Services via an Anthropic API Key or where we otherwise explicitly permit it, to access the Services through automated or non-human means, whether through a bot, script, or otherwise. So it seems like using a harness or your own tools to cal…

What is the point of a cli if you aren’t allowed to script it? Nonsense.

Re: Claude Code Routines

#289

Post by alex_sf »

crystal_revenge wrote: post #153
joshstrange wrote: post #66

LLMs and LLM providers are massive black boxes. I get a lot of value from them and so I can put up with that to a certain extent, but these new "products"/features that Anthropic are shipping are very unappealing to me. Not because I can't see a use-case for them, but because I have 0 trust in them: - No trust that they won't nerf the tool/model behind the feature - No trust they won't sunset the feature (the graveya…

This sounds like someone complaining about how Windows is a black box while ignoring the existence of Linux/BSD. I'm currently hosting, on very reasonable consumer grade hardware, an LLM that is on par performance wise what every anyone was paying for about a year ago. Including all the layers in between the model and the user. Llama.cpp serves up Gemma-4-26B-A4B, Open WebUI handles the client details: system prompt,…

Everytime I've tried a local model, and I have tried lots for a couple years now, they just seem like they were overtrained on benchmarks. They consistently perform dramatically worse than even older models from Anthropic/OAI/Google.

Re: Claude Code Routines

#290

Post by weird-eye-issue »

blcknight wrote: post #264
weird-eye-issue wrote: post #261

Earlier quoted context omitted.

--resume works fine?

That's why I mentioned `-p`. `--continue` and `--resume` are broken from `-p` sessions for the last 2 weeks. The use case is: 1. Do autonomous claudey thing (claude -p 'hey do this thing') 2. Do a deterministic thing 3. Reinvoke claude with `--continue` This no longer works. I've had this workflow in GitHub actions for months and all of a sudden they broke it. They constantly break stuff I rely on. Skill script loadi…

Been working fine for me with -p
Post reply on HN