Agents work better with small cached context buffers

Original: Agents often externalize some context: a repository in coding agents, a corpus in RAG, and the user prompt in an RLM.

Source: x.com ↗