https://www.hansking.cn/blog A posts created with Next.js and Tailwind.css en-us admin@hansking.cn (HANS) admin@hansking.cn (HANS) Wed, 24 Jun 2026 10:00:00 GMT https://www.hansking.cn/blog/2026/AI-xi-tong-de-SLO-zhi-liang-yan-chi-ke-yong-xing-yu-an-quan https://www.hansking.cn/blog/2026/AI-xi-tong-de-SLO-zhi-liang-yan-chi-ke-yong-xing-yu-an-quan [{"value":"AI 系统的 SLO：质量、延迟、可用性与安全","url":"#ai-系统的-slo质量延迟可用性与安全","depth":1},{"value":"一、先分清 SLI、SLO 和 SLA","url":"#一先分清-slislo-和-sla","depth":2},{"value":"二、AI 系统至少需要四类 SLO","url":"#二ai-系统至少需要四类-slo","depth":2},{"value":"1. 质量 SLO","url":"#1-质量-slo","depth":3},{"value":"2. 延迟 SLO","url":"#2-延迟-slo","depth":3},{"value":"3. 可用性 SLO","url":"#3-可用性-slo","depth":3},{"value":"4. 安全 SLO","url":"#4-安全-slo","depth":3},{"value":"三、质量指标要有采样和标签","url":"#三质量指标要有采样和标签","depth":2},{"value":"四、错误预算让团队敢于做改进","url":"#四错误预算让团队敢于做改进","depth":2},{"value":"五、降级要提前设计，不要故障时临时发挥","url":"#五降级要提前设计不要故障时临时发挥","depth":2},{"value":"六、发布门禁要同时看离线和在线","url":"#六发布门禁要同时看离线和在线","depth":2},{"value":"七、告警要连接到行动","url":"#七告警要连接到行动","depth":2},{"value":"八、SLO 也要防止被滥用","url":"#八slo-也要防止被滥用","depth":2},{"value":"总结：SLO 是团队对用户体验的共同承诺","url":"#总结slo-是团队对用户体验的共同承诺","depth":2}] Wed, 24 Jun 2026 10:00:00 GMT admin@hansking.cn (HANS) AILLMSLO可观测性可靠性 https://www.hansking.cn/blog/2026/Agent-shi-bai-mo-shi-xun-huan-huan-jue-gong-ju-wu-yong-yu-shi-kong https://www.hansking.cn/blog/2026/Agent-shi-bai-mo-shi-xun-huan-huan-jue-gong-ju-wu-yong-yu-shi-kong [{"value":"Agent 失败模式：循环、幻觉、工具误用与失控","url":"#agent-失败模式循环幻觉工具误用与失控","depth":1},{"value":"一、无限循环：任务在动，结果不动","url":"#一无限循环任务在动结果不动","depth":2},{"value":"二、状态漂移：模型记住的和系统发生的不是一回事","url":"#二状态漂移模型记住的和系统发生的不是一回事","depth":2},{"value":"三、幻觉参数：工具调用合法，含义却是假的","url":"#三幻觉参数工具调用合法含义却是假的","depth":2},{"value":"四、工具误用：模型选择了错误能力","url":"#四工具误用模型选择了错误能力","depth":2},{"value":"五、权限越界：最不能靠 Prompt 解决","url":"#五权限越界最不能靠-prompt-解决","depth":2},{"value":"六、无止境重试：把故障放大成事故","url":"#六无止境重试把故障放大成事故","depth":2},{"value":"七、计划漂移：目标在执行中悄悄改变","url":"#七计划漂移目标在执行中悄悄改变","depth":2},{"value":"八、如何设计止损机制","url":"#八如何设计止损机制","depth":2},{"value":"九、把失败变成可观察事件","url":"#九把失败变成可观察事件","depth":2},{"value":"十、用故障注入测试 Agent","url":"#十用故障注入测试-agent","depth":2},{"value":"总结：失控不是模型突然变坏，而是边界没有落地","url":"#总结失控不是模型突然变坏而是边界没有落地","depth":2}] Sun, 20 Jul 2025 10:00:00 GMT admin@hansking.cn (HANS) AILLMAgent可靠性安全工程 https://www.hansking.cn/blog/2026chang-ren-wu-Agent-jian-cha-dian-hui-fu-yu-mi-deng-zhi-xing https://www.hansking.cn/blog/2026chang-ren-wu-Agent-jian-cha-dian-hui-fu-yu-mi-deng-zhi-xing [{"value":"长任务 Agent：检查点、恢复与幂等执行","url":"#长任务-agent检查点恢复与幂等执行","depth":1},{"value":"一、先把任务从对话里拿出来","url":"#一先把任务从对话里拿出来","depth":2},{"value":"二、用状态机表达任务生命周期","url":"#二用状态机表达任务生命周期","depth":2},{"value":"三、检查点要保存什么","url":"#三检查点要保存什么","depth":2},{"value":"四、最危险的是“成功了，但我没收到成功响应”","url":"#四最危险的是成功了但我没收到成功响应","depth":2},{"value":"五、重试不是遇到错误就再来一次","url":"#五重试不是遇到错误就再来一次","depth":2},{"value":"六、恢复时不要盲目重跑整个计划","url":"#六恢复时不要盲目重跑整个计划","depth":2},{"value":"七、进度要让用户看得懂","url":"#七进度要让用户看得懂","depth":2},{"value":"八、检查点本身也要版本化和保护","url":"#八检查点本身也要版本化和保护","depth":2},{"value":"九、可靠性评测要模拟真实故障","url":"#九可靠性评测要模拟真实故障","depth":2},{"value":"总结：长任务 Agent 的记忆应该属于系统","url":"#总结长任务-agent-的记忆应该属于系统","depth":2}] Sat, 06 Jun 2026 10:00:00 GMT admin@hansking.cn (HANS) AILLMAgent工作流引擎可靠性