Hacker News Reader

edwinarbuson Claude Code regularly sends each background agent’s transcript to its model

I work on Claude Code. You've got the basic mechanism right, but to be super clear, this only applies to background agents that are actively working. They get a summary call about every 30 seconds, and idle background agents don't. These summaries are used for both you and Claude Code to track what each background agent is working on, without reading transcripts.

Your 16.4% sounds like it was for a team running mostly background agents, and doesn't seem like a typical session. (Feel free to send me the session ID at edwinarbus@anthropic.com and we can dig further.)

Last week, across Claude Code users who send us telemetry, agent_summary was about 2% of cost at API list prices and just under 4% of requests. In the terminal, where nearly all of these calls happen, it was about 6% of cost.

That said, we agree that's a lot for a status line. We just rolled out a fix for the most expensive case, summaries that miss the prompt cache (2.1.311+), and right now we're working on a much cheaper mode that sends short notes to a smaller model instead of the full transcript.

edwinarbuson Claude Code’s suggested message feature: I think the real customer is the model

I work on Claude Code. Prompt suggestions aren't being used to collect preference signals. We built this feature purely to help you stay in the flow, or for people who are returning to the session after a while and may need a little reminder on what a next step could be. We do see how many suggestions are accepted, but that's only so we know how helpful this feature is overall! (We tried to strike a balance by not making it "autocorrect", so it's more of a grayed out suggestion that can be typed over. It can always be toggled on/off in settings if you don't like it.)

edwinarbuson GPT-5

cross-posting:

https://x.com/sama/status/1953513280594751495 "wow a mega chart screwup from us earlier--wen GPT-6?! correct on the blog though."

blog: https://openai.com/index/introducing-gpt-5/

edwinarbuson OpenAI Audio Models

Just to confirm, is it this sample under Connection details? https://platform.openai.com/docs/guides/realtime?text-genera...

edwinarbuson New tools for building agents

sorry, fixed link: https://openai.github.io/openai-agents-python/

edwinarbuson OpenAI threatens to revoke o1 access for asking it about its chain of thought

Edwin from OpenAI here. 1) The linked tweet shows behavior through ChatGPT, not the OpenAI API, so you won't be charged for any tokens. 2) For the overall flow and email notification, we're taking a second look here.

edwinarbuson Flurly has been shut down by Stripe

(Edwin from Stripe here.) This is a really unfortunate situation. I can't talk specifics since it's a regulatory/legal matter and the decision wasn't up to us. But as others have pointed out, Stripe Connect should be used if you're building a marketplace. This gives the marketplace owner, Stripe, and the card networks some visibility into what's being exchanged—and so the marketplace owner can put proper safeguards in place to keep the marketplace safe, for both sellers and their customers. I'm saddened by what happened and I hope this does not deter JR from building future products.

edwinarbuson Stripe Financial Connections

Edwin from Stripe here. Stripe does integrate directly with banks. In our beta period, most volume we’ve seen has been over bank APIs. Some banks do not have APIs—we use financial partners to connect with them, and we’re talking with many banks in hopes that they will enable direct API access soon.

Our pricing is upfront: https://stripe.com/en-us/financial-connections#pricing. We’ve worked with a large beta group of users to make sure the pricing is in line with what they see in the market.

edwinarbuson Stripe's API was down

We're back up as of 17:02 UTC: https://twitter.com/stripestatus/status/1149002362691833856