Claude Code Routines

Post by causal » Wed, Apr 15, 2026, 1:07 PM UTC

mrgill wrote: post #385

andai wrote: post #25

I'm a little confused on the ToS here. From what I gathered, running `claude -p ` on cron is fine, but putting it in my Telegram bot is a ToS violation (unless I use per-token billing) because it's a 3rd party harness, right? (`claude -p` being a trivial workaround for the "no 3rd party stuff on the subscription" rule) This Routines feature notably works with the subscription, and it also has API callbacks. So if my…

Don't use claude -p in any kind of harness at all. I used it ONCE in a local custom made one, and got my account nuked. No help from them at all. Appealed, and the appeal was denied. I had been a Max subscriber since day 1 and Claude subscriber since day 1 in 2023. They don't care.

Post by Jimpulse » Wed, Apr 15, 2026, 1:33 PM UTC

dacox wrote: post #90

giancarlostoro wrote: post #86

Earlier quoted context omitted.

I've been talking to friends about this extensively, and read all sorts of different social media posts on X where people deep dove things (I'm at work so I don't have any links handy - though I did submit one on HN, grain of salt, unsure how valid it is but it was interesting: https://news.ycombinator.com/item?id=47752049 ). I think the real issue stems from the 1 Million token context window change. They did not an…

Yeah, I have been seeing lots of comments, tweets, etc, but given everything I have learned about these models - i do not think the change to 1M was innocuous. I'm not sure what they've claimed publicly, but I'm fairly certain they must be doing additional quantization, or at minimum additional quantization of the KV cache. Plus, sequence length can change things even when not fully utilized. I had to manually re-ena…

Post by rbalicki » Wed, Apr 15, 2026, 2:13 PM UTC

jplusequalt wrote: post #404

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…

>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. Or, you know, by writing the code yourself?

Post by pc86 » Wed, Apr 15, 2026, 2:46 PM UTC

jplusequalt wrote: post #404

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…

>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. Or, you know, by writing the code yourself?

Post by tomerhamam » Wed, Apr 15, 2026, 3:01 PM UTC