Improved token efficiency for longer agent runs
Cursor cut agent token cost by shrinking static instructions, loading tools on demand, improving cache boundaries, and using subagents more selectively.
Cursor reports **7% lower user token costs** without reduced agent quality after trimming its system prompt by about **66%**, cutting built-in tool descriptions in static context by **60%**, and reducing cold cache misses by **20%**.
Harness builders should keep only frequently needed tools in static context, place variable setup after stable cache boundaries, reduce incidental file-read overhead, and delegate when context isolation outweighs coordination cost.
Cursor reports **7% lower user token costs** without reduced agent quality after trimming its system prompt by about **66%**, cutting built-in tool descriptions in static context by **60%**, and reducing cold cache misses by **20%**. Harness builders should keep only frequently needed tools in static context, place variable setup after stable cache boundaries, reduce incidental file-read overhead, and delegate when context isolation outweighs coordination cost. These results come from Cursor production traffic and A/B tests, so they may not transfer directly to another request mix. The post also warns that subagents can duplicate work when isolated contexts lack coordination.
This adds production A/B evidence that substantial harness savings can come from shrinking static instructions, tool schemas, and cache misses without changing models. It reinforces deliberate compaction and selective context loading, while narrowing the case for subagents: isolation can save context, but poor coordination may duplicate work and erase those gains.