📄 中文摘要
AI 代理在运行过程中，随着时间的推移，所依赖的上下文信息可能变得陈旧、矛盾和冗余，导致决策质量下降。大多数团队关注代理的记忆内容，而忽视了其持续记忆的影响。为了改善这一问题，提出了定期遗忘的策略，强调信号与噪声的管理。通过每周的归档旧日志等步骤，可以有效提升代理的决策能力，确保其优化基于最新的信息，而非过时的背景。此方法不仅关注删除信息，更注重如何保持有效的知识管理。
记忆策划规则：为何你的 AI 代理需要定期遗忘
📄 English Summary
The Memory Curation Rule: Why Your AI Agent Needs to Forget on a Schedule
AI agents often suffer from decision-making deterioration over time due to reliance on outdated, contradictory, and bloated context. Most teams focus on what the agent remembers rather than what it continues to remember. To address this issue, a scheduled forgetting strategy is proposed, emphasizing signal-to-noise discipline. Implementing a weekly pattern, such as archiving old logs, can significantly enhance the decision-making capabilities of the agent, ensuring it optimizes based on the most current information rather than outdated contexts. This approach prioritizes effective knowledge management over mere deletion of information.