About this skill
面向AI Agent的精英级长期记忆系统,解决跨会话遗忘、检索不准、成本失控三大核心痛点. WAL写前日志协议采用"先写状态,再回复用户"模式,在压缩/崩溃/重启场景下保证上下文不丢失. 向量+关键词+图谱三路混合检索,通过RRF融合排序解决单一向量检索"相似但不相关"的召回质量问题. 六层存储架构(L1热内存→L2温向量→L3冷图谱→L4精选归档→L5云备份→L6自动抽取)按需分层加载,降低token消耗. 成本预算控制通过embedding缓存、分层存储、小模型抽取、图谱增量更新四项策略,配置日/月预算上限与自动降级. 记忆卫生自动化执行去重、衰减、晋升、降级、归档、删除六项维护,避免记忆膨胀拖慢召回. 适用于跨会话项目开发、多代理协作、长期偏好维护等生产级Agent记忆管理场景.
Quick start
- Copy the install prompt. Use the green button above this README.
- Send it to your agent. Paste it into your coding agent’s chat and send.
- Review the setup. Your agent checks the skill’s instructions, requirements, and installation scope.
Review, license & listing details
This is an imported listing. We haven’t certified its safety, license, or compatibility. Skills may include executable scripts.
HTTP 404
About scan coverage and limitationsAbout this listing
- Category
- General
- Reported license
- Not specified
- Listing language (estimated)
- Chinese
- Review status
- Not individually reviewed
- Agent compatibility
- Atlas installation supports Claude Code, Codex, and Cursor. Skill-specific dependencies and features are checked during setup.