Super curious to hear feedback or suggestions. We have a lot of different directions we're going to explore, the next big things we're looking at are:
Model Router - Routing between different models seamlessly without killing cache rate
Incremental Progress - Add the ability for the AI system to tweak tools/prompts seamlessly to improve cost/efficiency/response
Skill / API Aggregation - Avoiding the MiTM route and using a skill so the AI reports its usage. The idea is that this would enable it to work with things like Claude Review, etc...
TreDub•53m ago
Super curious to hear feedback or suggestions. We have a lot of different directions we're going to explore, the next big things we're looking at are:
Model Router - Routing between different models seamlessly without killing cache rate
Incremental Progress - Add the ability for the AI system to tweak tools/prompts seamlessly to improve cost/efficiency/response
Skill / API Aggregation - Avoiding the MiTM route and using a skill so the AI reports its usage. The idea is that this would enable it to work with things like Claude Review, etc...