When we looked at what our agent was sending, most of it wasn't the question. It was search results, tool output and old messages the model didn't need for that step.
On our demo agent (5 questions):
Before: 52.6K tokens per run, $0.129
After trimming: 9.1K tokens, $0.057
83% fewer tokens sent to the model
Nothing about the model or the prompt changed. We only stopped sending context that didn't matter for the task.
Your LLM bill grows with context the model never needed: extra search results, raw tool output, old messages. Contextwaise trims it before each call. Prompt-cache safe, every removal logged, trimmed parts one call away. ~100 ms, any provider.
