shreyas8

5 ms·

1. 1.▲
   by shreyas8 7mo ago

   Spot on, realtime cost visibility changes behavior more than any hard limit does. On CLI latency with concurrent users,each request spawns its own process so they don't block each other, but yeah, 3-8s per call isn't great at scal
2. 2.▲
   by shreyas8 7mo ago

   Yup, it likely is — I state this out in the README disclaimer. Built it before Anthropic's Feb 2026 policy update, discovered the TOS issue while prepping to open-source. Published it anyway since the per-key cost controls and usage tr
3. 3.▲
   Show HN: Bridge your Claude/OpenAI subs into a team API with per-key cost caps(github.com)

   1 points by shreyas8 7mo ago|4 comments