AWS Explores Prompt Caching Patterns for AI Agents on Bedrock
The AWS Machine Learning Blog explains how Amazon Bedrock prompt caching can reuse previously processed context, including instructions, tool definitions, and conversation history, instead of resending the full context on every agent step. The guide covers implicit and explicit cache breakpoints, stable prefixes, additional tools, cache eligibility, and write costs. In one six-turn workload, AWS reported about 62% lower input cost with implicit caching and about 61% with explicit rolling breakpoints on GPT-6 Astra, while cautioning that results vary by workflow.
AWS says the patterns can reduce repeated input processing while preserving agent continuity, but the reported savings depend on workload and cache reuse.
Every time you spin off your agent, it has to re-read its whole life story before performing an action. Here’s how to make it remember without having to feed it the same context again.
In the movie Groundhog Day, Bill Murray’s character re-lives the same February 2 every day. He wakes up to the same song playing on the radio, in the same inn, and the same small town, again and again for what some movie fans estimate to be 34 years.
Nothing he accomplished the day before carries over unless he works it out again from scratch. AI agents live in a version of Groundhog Day. Every time an agent takes a step (calls a tool, reads the result, decides what to do next), it has to send the model its entire story so far to maintain continuity: the tool definitions, the instructions, every previous tool call, and every result. The model only predicts the next token based on the input you provide. It has no internal memory between requests, so it must read all of it again. By turn 20 of a coding session, the model re-reads turns 1–19, and it’s you who pays for every one of those tokens, again and again.
#Amazon Bedrock#AWS Machine Learning Blog#OpenAI#GPT-6 Astra#GPT-6.1 Sol#GPT-5.6 Luna#Amazon Bedrock Runtime#Amazon Bedrock Mantle