扫码关注官方微信
扫码下载APP
发现最适合你需求的 AI 技能
|
定款商品原图无损下载&规范归档 - 人工触发单款操作，100% 保留原图，侵权风险标注，完整性校验
DEFAULT OUTPUT MODE: Always emit machine-parseable `openclaw-process` fenced JSON blocks in your assistant reply so a custom web client can render a live progress panel. Use when: any user message. Skip ONLY when the user explicitly requests no intermediate process (e.g. '只给最终答案'). Keep it lightweight for simple Q&A.
［何时使用］当用户需要系统化定义问题、设定成功标准、识别风险时；当用户说"帮我分析这个问题"时；当面临重大决策/战略模糊/复杂情境时；当需要将模糊问题转化为清晰行动时
主动触发引擎v4：基于兴趣衰减、沉默检测、话题热度，决定何时进行主动干预。触发条件：用户沉默、兴趣衰减、主动干预时机成熟。
自动捕获经验并安全进化的技能。触发条件：(1)命令/操作失败时→记ERRORS.md (2)被用户纠正('不对'/'应该是')时→记LEARNINGS.md (3)用户需要不存在的能力时→记FEATURE_REQUESTS.md (4)外部API/工具出错时→记ERRORS.md (5)发现自己知识过时/错误时→记LEARNINGS.md (6)发现更好做法时→记LEARNINGS.md (7)每个任务完成时→回顾过程，有新经验则记LEARNINGS.md。去重原则：如果没有新经验或已有条目已覆盖则跳过不写。每次写入同时在.learnings/CHANGELOG.md追加JSONL日志。经验反复出现≥3次时晋升到AGENTS.md/TOOLS.md/SOUL.md。详见正文。
|-
Proactive operations monitoring for OpenClaw agents. Tracks token utilization, memory layer health, and generates alerts. Provides `/health` dashboard and auto-suggestions for open loops. Works with memory-stack-core to prevent context overflow.
Transform AI agents from task-followers into proactive partners that anticipate needs and continuously improve. Now with WAL Protocol, Working Buffer, Autonomous Crons, and battle-tested patterns. Part of the Hal Stack 🦞"
Transform AI agents from task-followers into proactive partners that anticipate needs and continuously improve. Now with WAL Protocol, Working Buffer, Autonomous Crons, and battle-tested patterns. Part of the Hal Stack 🦞
Archiver·手机版·闲社网·闲社论坛·羊毛社区· 多链控股集团有限公司 · 苏ICP备2025199260号-1
Powered by Discuz! X5.0 © 2024-2025 闲社网·线报更新论坛·羊毛分享社区·http://xianshe.com
