Agent cost per successful task: a Zenn design-variable argument

The article positions agent cost as something decided at design time, not something fixed by shortening prompts. It focuses on three points: cache behavior determined by prefix stability (tools → system → messages), routing judged by a break-even formula, and budget enforcement in the application layer because AWS Budgets updates at most three times a day with 8–12 hour intervals. It also notes that input_tokens (inputTokens on Bedrock) only counts the portion not subject to caching, so total input must be aggregated with cache read and write. A concrete calculation for a 10-turn task on Sonnet 5 gives 0.146 for hits and $0.519 for constant misses. The thought is that once hits are working, output accounts for about one-third of cost, making the output side the next lever. One notable risk raised is that repeatedly missing with delimiters in place means paying 1.25x the write unit price continuously, which can end up costing more than no caching at all.