Yeah, the main point here is they had a CLI specifically that allowed you to call Claude, and that was being used. The CLI giving you access should kind of indicate that you should be able to use it as it is defined in the help.
I do agree, though, that the parts of this that were actually using the Claude system to generate OAuth keys themselves are a little sus.
That makes sense to say “must use Claude harness to login before calling Claude cli or using Claude code sdk”
v0 is moving from message-based billing to usage-based billing with tokens. Starting at your next billing cycle, your usage will be measured in input and output tokens and pricing will be more transparent, displayed in dollars.
Perhaps relevant that npm has been having issues although they only recently caught and fixed them. Scoped Private npm packages were getting cloudflare 503 errors
I authored the post and yes you are correct about the AWS Lambda being another service. I have since modified it based on a users suggestion to reflect.
I guess the point was that by adding the middleman between it there was more control --- We are self-funded and small group of people. We can't afford to hire a bunch of people to know about containers and servers and every little detail.
Thanks for your comments! Didn't expect so many views so quickly on this...
I do agree, though, that the parts of this that were actually using the Claude system to generate OAuth keys themselves are a little sus.
That makes sense to say “must use Claude harness to login before calling Claude cli or using Claude code sdk”