
  <rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
    <channel>
      <title>TechLog</title>
      <link>https://joker1point.github.io/techlog/blog</link>
      <description>AI Agent 工程化实践复盘：架构设计、容错与可靠性、上下文工程与成本治理。</description>
      <language>zh-cn</language>
      <managingEditor>birenwenmou@qq.com (眠月)</managingEditor>
      <webMaster>birenwenmou@qq.com (眠月)</webMaster>
      <lastBuildDate>Tue, 15 Sep 2026 00:00:00 GMT</lastBuildDate>
      <atom:link href="https://joker1point.github.io/techlog/tags/agent/feed.xml" rel="self" type="application/rss+xml"/>
      
  <item>
    <guid>https://joker1point.github.io/techlog/blog/agent-architecture-evolution-skill-restatement</guid>
    <title>我的 Agent 架构演进：Skill 压平 Sub Agent，Restatement 又把它拉了回来</title>
    <link>https://joker1point.github.io/techlog/blog/agent-architecture-evolution-skill-restatement</link>
    <description>从初代「主 Agent 编排 + 子 Agent 分工」三层架构，到用 Skill 替代冗余 Sub Agent 压平架构降低 Token，再到长任务中撞上步骤合并跳过、规则失效、输出同质化，最终靠 Restatement 机制与 Sub Agent 回归分头解决——一条完整的踩坑迭代链。</description>
    <pubDate>Tue, 15 Sep 2026 00:00:00 GMT</pubDate>
    <author>birenwenmou@qq.com (眠月)</author>
    <category>agent</category><category>ai</category><category>architecture</category><category>context-engineering</category>
  </item>

  <item>
    <guid>https://joker1point.github.io/techlog/blog/agent-fault-tolerance-design</guid>
    <title>Agent 健壮性设计：工具调用与模型调用失败的容错与兜底</title>
    <link>https://joker1point.github.io/techlog/blog/agent-fault-tolerance-design</link>
    <description>错误分类（确定性 vs 瞬时）→ 确定性错误回传模型自我修复、瞬时错误指数退避+随机抖动重试 → 多次失败熔断 → 按能力等级优雅降级（备用工具 / 业务降级 / 转人工）。模型侧另需识别「API 返回成功但业务结果错误」的潜在失败，并在切换备用模型后持续评估。</description>
    <pubDate>Tue, 15 Sep 2026 00:00:00 GMT</pubDate>
    <author>birenwenmou@qq.com (眠月)</author>
    <category>agent</category><category>ai</category><category>reliability</category><category>system-design</category>
  </item>

  <item>
    <guid>https://joker1point.github.io/techlog/blog/agent-lessons-precipitation</guid>
    <title>如何沉淀 Agent 每次的踩坑经验</title>
    <link>https://joker1point.github.io/techlog/blog/agent-lessons-precipitation</link>
    <description>把&quot;记住教训&quot;从 agent 的自觉变成基础设施：三层漏斗体系——规则强制捕获（项目级原始档）→ 定期蒸馏（全局技能）→ 协议化进化（活性维护）。</description>
    <pubDate>Sun, 06 Sep 2026 00:00:00 GMT</pubDate>
    <author>birenwenmou@qq.com (眠月)</author>
    <category>agent</category><category>dev-lessons</category><category>engineering</category><category>methodology</category><category>knowledge-management</category>
  </item>

  <item>
    <guid>https://joker1point.github.io/techlog/blog/agent-memory-safe-compaction</guid>
    <title>AI 记忆的抗膨胀：从&quot;淘汰谁&quot;到&quot;留下谁&quot;</title>
    <link>https://joker1point.github.io/techlog/blog/agent-memory-safe-compaction</link>
    <description>跨会话的项目记忆文件会像所有&quot;活&quot;的文档一样膨胀。本文记录一次从 26KB 到 14.7KB 的安全精简：为什么&quot;热度评分淘汰&quot;不可行（使用不可观测、低频高损会误杀），改成&quot;准入三条&quot;，以及&quot;先归档 → 先落点 → 后删除 → 再对账&quot;的搬迁式精简流程与两个对账脚本。</description>
    <pubDate>Fri, 09 Oct 2026 00:00:00 GMT</pubDate>
    <author>birenwenmou@qq.com (眠月)</author>
    <category>agent</category><category>memory</category><category>knowledge-management</category><category>engineering</category>
  </item>

  <item>
    <guid>https://joker1point.github.io/techlog/blog/agent-prompt-injection-defense</guid>
    <title>Agent 怎么防止 Prompt 注入：四层纵深防御体系</title>
    <link>https://joker1point.github.io/techlog/blog/agent-prompt-injection-defense</link>
    <description>单靠 Prompt 加限制词防不住注入，正确做法是搭建「输入侧 → 模型侧 → 执行侧 → 输出侧」四层纵深防御体系，其中执行侧权限管控是最硬的底线。</description>
    <pubDate>Sun, 06 Sep 2026 00:00:00 GMT</pubDate>
    <author>birenwenmou@qq.com (眠月)</author>
    <category>ai</category><category>agent</category><category>security</category><category>prompt-injection</category><category>llm</category>
  </item>

  <item>
    <guid>https://joker1point.github.io/techlog/blog/agent-skill-governance</guid>
    <title>我是如何保证 Agent Skill 多而不乱的？</title>
    <link>https://joker1point.github.io/techlog/blog/agent-skill-governance</link>
    <description>700+ Skill 库的四层治理体系：L1 排序层（选得对）、L2 存储管理层（管得住）、L3 分区保护层（改不坏）、L4 自进化层（不腐烂），叠加快慢双路径检索。</description>
    <pubDate>Sun, 06 Sep 2026 00:00:00 GMT</pubDate>
    <author>birenwenmou@qq.com (眠月)</author>
    <category>agent</category><category>skill</category><category>governance</category><category>engineering</category><category>methodology</category>
  </item>

  <item>
    <guid>https://joker1point.github.io/techlog/blog/agent-washing-why-no-value</guid>
    <title>Agent Washing：为什么大批企业升级 Agent 不会产生增值</title>
    <link>https://joker1point.github.io/techlog/blog/agent-washing-why-no-value</link>
    <description>Agent 有明确适用边界——擅长非结构化输入、路径不固定、需动态推理的任务。把确定流程强行套上自主智能是技术错配；中小企业缺数据/SOP/权限/复合人才配套地基；管理层&quot;替换人&quot;的预期错位（实际只能替换重复与信息整理类子集工作）。</description>
    <pubDate>Tue, 15 Sep 2026 00:00:00 GMT</pubDate>
    <author>birenwenmou@qq.com (眠月)</author>
    <category>agent</category><category>ai</category><category>product</category><category>industry</category>
  </item>

  <item>
    <guid>https://joker1point.github.io/techlog/blog/agent-window-period</guid>
    <title>对Agent产品窗口期的看法</title>
    <link>https://joker1point.github.io/techlog/blog/agent-window-period</link>
    <description>Agent的窗口期会非常短——因为Agent本身就在提高开发Agent应用的效率，市场需求将迅速饱和。以月之暗面Alnative桌面级本地Agent为例，超5万行代码中92%由AI自主生成，工程师仅耗时一周完成架构优化与交互调校。</description>
    <pubDate>Sat, 06 Jun 2026 00:00:00 GMT</pubDate>
    <author>birenwenmou@qq.com (眠月)</author>
    <category>ai</category><category>agent</category><category>startup</category><category>product</category><category>opinion</category>
  </item>

  <item>
    <guid>https://joker1point.github.io/techlog/blog/ai-eleven-defenses</guid>
    <title>让 AI 写代码背后的十一道防线</title>
    <link>https://joker1point.github.io/techlog/blog/ai-eleven-defenses</link>
    <description>一份脱敏的真实项目复盘：用 69 份文档、1067 条标题、560 份验证留痕，在 AI 动手写代码之前搭出十一道防线——目的不是限制 AI，而是敢让它碰生产系统。</description>
    <pubDate>Wed, 14 Oct 2026 00:00:00 GMT</pubDate>
    <author>birenwenmou@qq.com (眠月)</author>
    <category>ai-coding</category><category>agent</category><category>engineering</category><category>dev-process</category><category>datawhale</category>
  </item>

  <item>
    <guid>https://joker1point.github.io/techlog/blog/dev-lessons-skill-distillation</guid>
    <title>把 22 篇项目文档蒸馏成一个可进化的全局技能</title>
    <link>https://joker1point.github.io/techlog/blog/dev-lessons-skill-distillation</link>
    <description>将 22 篇原始复盘文档蒸馏为全局技能 dev-lessons：按领域分文件（LLM 应用、前后端异步、数据库迁移、上游复用、部署环境、协作测试），每条经验附场景标签与溯源字段，并内置进化协议（同问题复现三次升级为元原则、元原则超 15 条强制合并）。</description>
    <pubDate>Tue, 15 Sep 2026 00:00:00 GMT</pubDate>
    <author>birenwenmou@qq.com (眠月)</author>
    <category>agent</category><category>skill</category><category>engineering</category><category>knowledge-management</category>
  </item>

  <item>
    <guid>https://joker1point.github.io/techlog/blog/github-trending-2026-09</guid>
    <title>2026 年 9 月 GitHub 热门项目分析：追踪大模型应用领域技术动态</title>
    <link>https://joker1point.github.io/techlog/blog/github-trending-2026-09</link>
    <description>2026 年 9 月 GitHub 月榜 Top 10 逐项拆解——从 archify（把 AI 生成的架构图变成可验证制品）到 OpenViking（Agent 上下文数据库），并归纳出 Agent 输出制品化、上下文工程、Rust 工具化、AI 与教育/桌面交叉四条趋势。</description>
    <pubDate>Thu, 17 Sep 2026 00:00:00 GMT</pubDate>
    <author>birenwenmou@qq.com (眠月)</author>
    <category>github</category><category>open-source</category><category>ai</category><category>agent</category><category>trending</category>
  </item>

  <item>
    <guid>https://joker1point.github.io/techlog/blog/ide-agent-vs-ai-assistant</guid>
    <title>IDE Agent 开发体验：为什么直接操作代码比通用 AI 助手更靠谱</title>
    <link>https://joker1point.github.io/techlog/blog/ide-agent-vs-ai-assistant</link>
    <description>对比 IDE 内置 Agent 与通用 AI 编程助手在代码开发场景中的体验差异，探讨代码硬约束、模型幻觉抑制、精准代码定位等关键优势。</description>
    <pubDate>Thu, 04 Jun 2026 00:00:00 GMT</pubDate>
    <author>birenwenmou@qq.com (眠月)</author>
    <category>ai</category><category>ide</category><category>agent</category><category>devtools</category><category>opinion</category><category>developer-experience</category>
  </item>

  <item>
    <guid>https://joker1point.github.io/techlog/blog/mpt-open-source-contribution-analysis</guid>
    <title>如何给一个陌生开源项目做架构分析，并找到能合进去的贡献点</title>
    <link>https://joker1point.github.io/techlog/blog/mpt-open-source-contribution-analysis</link>
    <description>以 MoneyPrinterTurbo 为例，按 7 个维度（任务规划、多 Agent 协作、上下文管理、Human-in-the-loop、评估框架、工具路由、Streaming）系统拆解一个陌生项目的架构缺陷，从中筛出 Top 3 贡献方向，并给出方案骨架与给维护者的英文沟通草稿。</description>
    <pubDate>Tue, 15 Sep 2026 00:00:00 GMT</pubDate>
    <author>birenwenmou@qq.com (眠月)</author>
    <category>open-source</category><category>agent</category><category>architecture</category><category>python</category>
  </item>

  <item>
    <guid>https://joker1point.github.io/techlog/blog/mpt-pr-1291-contribution-review</guid>
    <title>给 119k star 项目提 PR：8 小时从评论到合并的完整复盘</title>
    <link>https://joker1point.github.io/techlog/blog/mpt-pr-1291-contribution-review</link>
    <description>一次向 MoneyPrinterTurbo（119k+ star）提交 OpenAI 兼容文生图素材源并被合并的完整复盘：占坑式沟通拿到约束、用数据回应&quot;改动过大&quot;的顾虑、测试证据链决定合并速度，以及维护者在后续 PR 中指出的付费 API 容错设计缺陷。</description>
    <pubDate>Tue, 15 Sep 2026 00:00:00 GMT</pubDate>
    <author>birenwenmou@qq.com (眠月)</author>
    <category>open-source</category><category>agent</category><category>python</category><category>engineering</category><category>pr</category>
  </item>

  <item>
    <guid>https://joker1point.github.io/techlog/blog/onboarding-project-deep-understanding</guid>
    <title>如何快速接手一个项目并构建深度项目认知</title>
    <link>https://joker1point.github.io/techlog/blog/onboarding-project-deep-understanding</link>
    <description>接手陌生项目的四阶段方法论：宏观定向（1-2小时）→ 架构建模（半天）→ 运行验证（半天）→ 持续深化。12 个可执行步骤，配合 agent 技能栈实现认知加速。</description>
    <pubDate>Sun, 06 Sep 2026 00:00:00 GMT</pubDate>
    <author>birenwenmou@qq.com (眠月)</author>
    <category>onboarding</category><category>engineering</category><category>architecture</category><category>methodology</category><category>agent</category>
  </item>

  <item>
    <guid>https://joker1point.github.io/techlog/blog/rag-vs-agent-retrieval-decision-framework</guid>
    <title>RAG 还必要吗：从 130 条技术评论里提炼出的选型决策框架</title>
    <link>https://joker1point.github.io/techlog/blog/rag-vs-agent-retrieval-decision-framework</link>
    <description>围绕「向量 RAG vs LLM Wiki + grep / Agent 自主检索」的四派立场梳理（谨慎派、混合派、场景决定论派、演进视角派）、四个辩论冲突点，以及一组可直接落地的决策框架：按更新频率、查询频率、数据规模、领域特性决定上不上 RAG。</description>
    <pubDate>Tue, 15 Sep 2026 00:00:00 GMT</pubDate>
    <author>birenwenmou@qq.com (眠月)</author>
    <category>rag</category><category>agent</category><category>ai</category><category>research</category><category>vdb-news</category>
  </item>

  <item>
    <guid>https://joker1point.github.io/techlog/blog/workflow-vs-agent-7-dimensions</guid>
    <title>Workflow vs Agent：用 7 维度框架定位你的项目</title>
    <link>https://joker1point.github.io/techlog/blog/workflow-vs-agent-7-dimensions</link>
    <description>以 MoneyPrinterTurbo 为例，用 7 个 agent 架构维度审视一个固定管线项目：它几乎一个都不占，但这不是缺陷，而是项目性质使然。能说清&quot;为什么这里不需要 agent&quot;，比硬说&quot;这里有 agent&quot;更显研究深度。</description>
    <pubDate>Sun, 06 Sep 2026 00:00:00 GMT</pubDate>
    <author>birenwenmou@qq.com (眠月)</author>
    <category>ai</category><category>agent</category><category>workflow</category><category>architecture</category><category>llm</category>
  </item>

    </channel>
  </rss>
