Cap prompt and output tokens before you tune quality. Use when an agent path is getting expensive or slow.
--- name: budget-tokens-per-request category: ai-agents description: >- Cap prompt and output tokens before you tune quality. Use when an agent path is getting expensive or slow. --- # Budget Tokens per Request Set a hard max for system + history + tools. Truncate oldest tool results first. Measure cost per successful task, not per call.
Shared publicly by the author and open for any use. See the Terms.