
  <rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
    <channel>
      <title>ApFramework</title>
      <link>https://apframework.com/blog</link>
      <description>Shoukai 技术博客，软件架构、AI、领域驱动、开源软件、阅读笔记</description>
      <language>zh</language>
      <managingEditor>huangshoukai@yeah.net (Shoukai Huang)</managingEditor>
      <webMaster>huangshoukai@yeah.net (Shoukai Huang)</webMaster>
      <lastBuildDate>Wed, 15 Apr 2026 00:00:00 GMT</lastBuildDate>
      <atom:link href="https://apframework.com/feed.xml" rel="self" type="application/rss+xml"/>
      
  <item>
    <guid>https://apframework.com/blog/essay/2026-04-15-cognee</guid>
    <title>Cognee 实战：构建 AI Agent 的可计算记忆层</title>
    <link>https://apframework.com/blog/essay/2026-04-15-cognee</link>
    <description>本文结合代码实践，解析 Cognee 如何将向量检索、来源追踪与知识图谱整合为面向 AI Agent 的可计算记忆层，并通过一个最小示例验证其本地接入与查询实践。</description>
    <pubDate>Wed, 15 Apr 2026 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>Cognee</category><category>RAG</category><category>GraphRAG</category><category>Knowledge Graph</category><category>LangChain</category><category>AI Agents</category><category>Memory</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2026-04-05-Superpowers</guid>
    <title>从 Vibe Coding 到 Viable Coding：Superpowers 工程化工作流</title>
    <link>https://apframework.com/blog/essay/2026-04-05-Superpowers</link>
    <description>本文系统拆解 Jesse Vincent 开源的 Superpowers 工作流，说明它如何通过 brainstorming、git worktree、writing-plans、subagent-driven development、TDD 与 code review 等门禁技能，把“先写再说”的 AI 编码升级为可设计、可验证、可审查、可合并的软件工程流程。</description>
    <pubDate>Sun, 05 Apr 2026 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>AI Coding</category><category>Agent</category><category>Superpowers</category><category>TDD</category><category>Spec Driven Development</category><category>Workflow</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2026-02-24-Human-Agent-Collaboration</guid>
    <title>构建 AI 智能体应用（八）：人机协作</title>
    <link>https://apframework.com/blog/essay/2026-02-24-Human-Agent-Collaboration</link>
    <description>本文是《Building Applications with AI Agents》系列解读的最后一篇。当 AI Agent 从工具进化为队友，如何设计人机协作模式？本文探讨从“执行者”到“治理者”的角色演变、Guided Autonomy 原则、多代理编排中的信任校准，以及 2026 年最新的治理与合规实践。</description>
    <pubDate>Tue, 24 Feb 2026 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>AI Agents</category><category>Human-Agent Collaboration</category><category>Guided Autonomy</category><category>Governance</category><category>Trust</category><category>Engineering</category><category>读书笔记</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2026-02-23-Protecting-Agentic-Systems</guid>
    <title>构建 AI 智能体应用（七）：保护系统安全</title>
    <link>https://apframework.com/blog/essay/2026-02-23-Protecting-Agentic-Systems</link>
    <description>本文是《Building Applications with AI Agents》系列解读的第七篇。随着 AI Agent 获得自主行动能力，安全边界从单纯的 Prompt 防护扩展到全链路工程治理。本文基于 2026 年最新视角，详解 OWASP 代理安全风险、MAESTRO 威胁建模框架以及多层防御体系的落地实践。</description>
    <pubDate>Mon, 23 Feb 2026 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>AI Agents</category><category>Security</category><category>OWASP</category><category>MAESTRO</category><category>Prompt Injection</category><category>Defense in Depth</category><category>Engineering</category><category>读书笔记</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2026-02-22-Improvement-Loops</guid>
    <title>构建 AI 智能体应用（六）：持续改进循环 (Improvement Loops)</title>
    <link>https://apframework.com/blog/essay/2026-02-22-Improvement-Loops</link>
    <description>本文是《Building Applications with AI Agents》系列解读的第六篇。AI Agent 的真正挑战在于“如何持续变好”。本文深入探讨反馈管道、实验验证与持续学习三大支柱，构建从生产洞察到能力进化的闭环体系。</description>
    <pubDate>Sun, 22 Feb 2026 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>AI Agents</category><category>Improvement Loops</category><category>Feedback Pipelines</category><category>Experimentation</category><category>Continuous Learning</category><category>Engineering</category><category>读书笔记</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2026-02-21-Monitoring-in-Production</guid>
    <title>构建 AI 智能体应用（五）：生产监控与可观测性实战</title>
    <link>https://apframework.com/blog/essay/2026-02-21-Monitoring-in-Production</link>
    <description>本文是《Building Applications with AI Agents》系列解读的第五篇。AI Agent 的生产监控不同于传统软件，需要同时覆盖基础设施信号与语义行为信号。本文提供从 OTel 技术栈选择到漂移检测、安全发布的完整实战指南。</description>
    <pubDate>Sat, 21 Feb 2026 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>AI Agents</category><category>Monitoring</category><category>Observability</category><category>OpenTelemetry</category><category>DevOps</category><category>SRE</category><category>读书笔记</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2026-02-19-Validation-Measurement</guid>
    <title>构建 AI 智能体应用（四）：验证与测量</title>
    <link>https://apframework.com/blog/essay/2026-02-19-Validation-Measurement</link>
    <description>本文是《Building Applications with AI Agents》系列解读的第四篇。AI Agent 的质量不等于“回答写得好”，而取决于它在真实环境里是否能稳定完成任务。本文提供从离线评估到生产监控的全链路落地指南，帮助工程团队把“看起来更聪明”的改动，变成“可证明更可靠”的改动。</description>
    <pubDate>Thu, 19 Feb 2026 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>AI Agents</category><category>Evaluation</category><category>Testing</category><category>Metrics</category><category>LLM-as-Judge</category><category>Best Practices</category><category>Engineering</category><category>读书笔记</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2026-02-17-multiagent-coordination</guid>
    <title>构建 AI 智能体应用（三）：多智能体协作与编排模式</title>
    <link>https://apframework.com/blog/essay/2026-02-17-multiagent-coordination</link>
    <description>本文是《Building Applications with AI Agents》系列解读的第三篇。从经典的顺序流水线到 2026 年前沿的联邦协作与 A2A 协议，深入解析多智能体系统（MAS）的四大核心维度：任务流转、质量控制、通信架构与标准化协议。</description>
    <pubDate>Tue, 17 Feb 2026 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>AI Agents</category><category>Multi-Agent Systems</category><category>Orchestration</category><category>MCP</category><category>A2A</category><category>架构演进</category><category>读书笔记</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2026-02-16-agent-tool-selection</guid>
    <title>构建 AI 智能体应用（二）：工具选择策略与 MCP 生态演进</title>
    <link>https://apframework.com/blog/essay/2026-02-16-agent-tool-selection</link>
    <description>本文是《Building Applications with AI Agents》系列解读的第二篇。继上一篇剖析智能体核心类型后，本篇深入探讨智能体能力的延伸——工具选择（Tool Selection）。从基础的 Prompt 工程到 2026 年主流的 Tool RAG 与 MCP 生态，为您揭示如何构建支持万级工具规模的高效 Agent 系统。</description>
    <pubDate>Mon, 16 Feb 2026 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>AI Agents</category><category>Tool Selection</category><category>MCP</category><category>RAG</category><category>Agent Skills</category><category>架构演进</category><category>读书笔记</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2026-02-15-agent-types</guid>
    <title>构建 AI 智能体应用（一）：核心架构模式演进</title>
    <link>https://apframework.com/blog/essay/2026-02-15-agent-types</link>
    <description>本文是《Building Applications with AI Agents》系列解读的第一篇。我们基于 Michael Albada 的系统化理论，深入解析 2026 年典型的 AI 智能体类型（从反射型到深度研究型），为您呈现智能体从单体应用向认知生态演进的全景图。</description>
    <pubDate>Sun, 15 Feb 2026 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>AI Agents</category><category>Agentic AI</category><category>架构演进</category><category>ReAct</category><category>反射型智能体</category><category>深度研究</category><category>多智能体编排</category><category>读书笔记</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2026-02-08-agentic-ai-architecture-framework</guid>
    <title>企业 Agentic AI 架构框架：2026 全景解析与落地指南</title>
    <link>https://apframework.com/blog/essay/2026-02-08-agentic-ai-architecture-framework</link>
    <description>本文是对 2025 年 InfoQ 框架的更新，融合了 2026 年的最佳实践和架构趋势。详细阐述了企业 Agentic AI 的三层架构（基础层、工作流程层、自治层），并引入 GraphRAG、代理 FinOps、深度可观测性及混合边缘/云架构等前沿概念，为企业实现智能自动化提供全面的架构指导。</description>
    <pubDate>Sun, 08 Feb 2026 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>Agentic AI</category><category>架构框架</category><category>企业级AI</category><category>GraphRAG</category><category>FinOps</category><category>混合智能</category><category>深度可观测性</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2026-01-29-langchain-multi-agent-skills</guid>
    <title>LangChain Skills 模式实战：构建按需加载知识的 SQL 助手</title>
    <link>https://apframework.com/blog/essay/2026-01-29-langchain-multi-agent-skills</link>
    <description>本文深入探讨如何利用 LangChain 的 Skills 模式构建高效 SQL Assistant。通过渐进式披露原则，解决传统 System Prompt 上下文过载问题，实现知识的按需加载。结合 Sales Analytics 和 Inventory Management 两个实际场景，详细演示从环境搭建、技能定义到 Agent 编排的完整流程。</description>
    <pubDate>Thu, 29 Jan 2026 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>LangChain</category><category>Skills</category><category>SQL</category><category>Agent</category><category>LangGraph</category><category>Python</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2026-01-18-content-extraction-LangGraph</guid>
    <title>知识提取：LangGraph 核心概念与内容精要</title>
    <link>https://apframework.com/blog/essay/2026-01-18-content-extraction-LangGraph</link>
    <description>本文利用 content-extraction 工具深度萃取 LangGraph 官方文档精华，旨在为智能体开发者构建结构化的知识底座。通过提炼核心概念与架构设计，帮助开发者快速掌握 LangGraph 全貌，打破信息茧房，避免重复造轮子。</description>
    <pubDate>Sun, 18 Jan 2026 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>LangGraph</category><category>Agent</category><category>Content Extraction</category><category>RAG</category><category>Knowledge Base</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2026-01-17-content-extraction-langchain</guid>
    <title>知识提取：LangChain 核心概念与内容精要</title>
    <link>https://apframework.com/blog/essay/2026-01-17-content-extraction-langchain</link>
    <description>本文利用 content-extraction 工具深度萃取 LangChain 官方文档精华，旨在为智能体开发者构建结构化的知识底座。通过提炼核心概念与架构设计，帮助开发者快速掌握 LangChain 全貌，打破信息茧房，避免重复造轮子。</description>
    <pubDate>Sat, 17 Jan 2026 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>LangChain</category><category>Agent</category><category>Content Extraction</category><category>RAG</category><category>Knowledge Base</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2026-01-10-HyDE-Multi-HyDE-Adaptive-HyDE</guid>
    <title>RAG 进阶：HyDE、Multi-HyDE 与 Adaptive HyDE 深度解析与实战</title>
    <link>https://apframework.com/blog/essay/2026-01-10-HyDE-Multi-HyDE-Adaptive-HyDE</link>
    <description>深入解析 HyDE 及其变体 Multi-HyDE 和 Adaptive HyDE，探讨如何利用大语言模型生成的假设文档提升 RAG 系统的检索准确性，并提供详细的代码实战与性能对比分析。</description>
    <pubDate>Sat, 10 Jan 2026 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>RAG</category><category>HyDE</category><category>Multi-HyDE</category><category>Adaptive HyDE</category><category>LLM</category><category>AI</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2026-01-02-LangChain-Multi-Agent-Swarm</guid>
    <title>LangChain 多智能体架构：深入解析 Swarm 的 Handoffs 模式</title>
    <link>https://apframework.com/blog/essay/2026-01-02-LangChain-Multi-Agent-Swarm</link>
    <description>本文深入解析 LangGraph Swarm 架构，通过核心原理拆解与源码级分析，揭示多智能体协作的底层机制。结合旅行预订系统实战案例，演示如何使用 Python 构建支持动态任务交接（Handoff）与状态共享的高效 AI 智能体集群，为构建企业级复杂 Agent 系统提供最佳实践参考。</description>
    <pubDate>Fri, 02 Jan 2026 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>LangChain</category><category>LangGraph</category><category>Swarm</category><category>Multi-Agent</category><category>Python</category><category>DeepSeek</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-12-24-LangChain-Deep-Agents-Claude-Skills-by-Sandbox</guid>
    <title>基于 LangChain Deep Agents 与 Daytona Sandbox 集成 Claude Skills 的智能体开发指南</title>
    <link>https://apframework.com/blog/essay/2025-12-24-LangChain-Deep-Agents-Claude-Skills-by-Sandbox</link>
    <description>本文详细介绍了如何构建一个具备深度思考和安全执行能力的智能体系统。通过整合 LangChain 的推理能力、Claude Skills 的动态技能机制以及 Daytona Sandbox 的安全执行环境，实现了从任务规划到代码执行的完整闭环。</description>
    <pubDate>Wed, 24 Dec 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>LangChain</category><category>AI Agents</category><category>Claude</category><category>Daytona</category><category>Python</category><category>LLM</category><category>ReAct</category><category>Sandbox</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-12-22-microsandbox</guid>
    <title>Microsandbox 体验：构建安全隔离的轻量级代码执行环境</title>
    <link>https://apframework.com/blog/essay/2025-12-22-microsandbox</link>
    <description>深入体验 Microsandbox：一个轻量级、硬件隔离的代码执行沙盒。本文详细介绍了其架构优势、安装配置流程。</description>
    <pubDate>Mon, 22 Dec 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>Microsandbox</category><category>Security</category><category>MCP</category><category>Python</category><category>DevOps</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-12-20-Anthropic-Claude-Skills</guid>
    <title>Anthropic Claude Skills 案例解析：构建高效 Agent 的核心指南</title>
    <link>https://apframework.com/blog/essay/2025-12-20-Anthropic-Claude-Skills</link>
    <description>深入解析 Anthropic 官方推出的 Skills 系统，探讨其作为 Agent 能力扩展的核心机制。本文以 PDF 处理技能为例，详细拆解如何通过渐进式上下文构建（Pyramid Principle）和模块化指令，提升 Claude 在专业领域的任务执行能力与可靠性。</description>
    <pubDate>Sat, 20 Dec 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>Anthropic</category><category>Claude</category><category>Agent</category><category>Python</category><category>PDF</category><category>Workflow</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-12-16-LangChain-Deep-Agents-Skills</guid>
    <title>Deep Agents 进阶：LangChain Deep Agents 集成 Claude Skills 技能指南</title>
    <link>https://apframework.com/blog/essay/2025-12-16-LangChain-Deep-Agents-Skills</link>
    <description>本文深入探讨如何在 LangChain Deep Agents 架构中集成 Skills 技能系统。文章对比了 CLI 与 Library 的实现差异，详细介绍了通过 CompiledSubAgent 封装技能、利用 Progressive Disclosure 动态加载以及 PythonREPLTool 执行环境的完整技术方案。</description>
    <pubDate>Tue, 16 Dec 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>LangChain</category><category>Deep Agents</category><category>Claude Skills</category><category>Python</category><category>Skills</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-12-14-Claude-Agent-Skills</guid>
    <title>Claude Agent Skills 深度解析：原理、工作流与最佳实践</title>
    <link>https://apframework.com/blog/essay/2025-12-14-Claude-Agent-Skills</link>
    <description>深入解析 Claude Agent Skills 功能，探讨其按需加载、可组合与代码执行的核心特性。本文详细解读 SKILL.md 结构、触发机制及最佳实践，助您构建更高效、更专业的 AI 智能体。</description>
    <pubDate>Sun, 14 Dec 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>Claude</category><category>Agent</category><category>Skills</category><category>AI Architecture</category><category>Anthropic</category><category>Engineering</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-12-09-LangChain-Multi-agent</guid>
    <title>LangChain 多智能体架构：深入解析 Supervisor 与 Handoffs 模式</title>
    <link>https://apframework.com/blog/essay/2025-12-09-LangChain-Multi-agent</link>
    <description>深入探讨 LangChain 多智能体架构的核心模式，从集中式的 Supervisor 到去中心化的 Handoffs。本文结合代码示例，解析如何构建具备工具治理、上下文感知与层级协作能力的复杂 AI 系统。</description>
    <pubDate>Tue, 09 Dec 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>LangChain</category><category>LangGraph</category><category>Multi-Agent</category><category>Supervisor</category><category>AI Architecture</category><category>Python</category><category>Agentic Patterns</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-12-01-Dify-Kubernetes</guid>
    <title>在 macOS 上用 Kind 部署 Dify 到 Kubernetes</title>
    <link>https://apframework.com/blog/essay/2025-12-01-Dify-Kubernetes</link>
    <description>面向开发者的本地 Kubernetes（Kind）快速环境搭建与 Dify 部署指南，覆盖工具准备、集群创建、常用运维工具、Helm 部署与问题排查。</description>
    <pubDate>Mon, 01 Dec 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>Kubernetes</category><category>Kind</category><category>Helm</category><category>Dify</category><category>macOS</category><category>DevOps</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-11-17-Agents-2</guid>
    <title>Agents 2.0：从浅层循环到深度智能（Deep Agents）</title>
    <link>https://apframework.com/blog/essay/2025-11-17-Agents-2</link>
    <description>浅层 Agent 依赖上下文窗、易陷循环且缺乏持久状态；深度 Agent 以规划/委派/持久记忆/人类在环为支柱，基于 LangGraph 的有状态编排与 DeepAgents harness，支撑小时到天级复杂任务。</description>
    <pubDate>Mon, 17 Nov 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>Agents 2.0</category><category>Agent</category><category>LangChain</category><category>LangGraph</category><category>Deep Agents</category><category>人类在环</category><category>持久化</category><category>长期记忆</category><category>子代理</category><category>AI工程化</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-10-29-chrome-devtools-mcp</guid>
    <title>Chrome DevTools MCP 实战指南：浏览器自动化解决方案</title>
    <link>https://apframework.com/blog/essay/2025-10-29-chrome-devtools-mcp</link>
    <description>深入解析 Chrome DevTools MCP 的核心架构与实现原理，详细介绍如何通过标准化的 MCP 协议为 AI 编程助手提供完整的 Chrome 浏览器控制能力。文章涵盖从环境配置、服务集成到实际应用的完整开发流程，并提供丰富的代码示例和 API 参考，帮助开发者快速构建高可靠性的浏览器自动化、性能分析和调试解决方案。</description>
    <pubDate>Wed, 29 Oct 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>Chrome DevTools</category><category>MCP</category><category>浏览器自动化</category><category>AI编程助手</category><category>Trae AI</category><category>Puppeteer</category><category>性能监控</category><category>调试工具</category><category>Web开发</category><category>JavaScript</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-10-25-DeepAgents</guid>
    <title>DeepAgents 深度解析：从概念到代码验证的完整指南</title>
    <link>https://apframework.com/blog/essay/2025-10-25-DeepAgents</link>
    <description>深入解析 DeepAgents 代理工具包的核心概念与技术架构。文章系统梳理了 DeepAgents、LangChain 和 LangGraph 三者的关系与定位，详细对比了代理工具包、代理框架和代理运行时的区别。通过完整的代码示例和详细的日志分析，展示了如何使用 DeepSeek 和 Tavily 构建智能研究代理，为开发者提供从理论到实践的完整学习路径。</description>
    <pubDate>Sat, 25 Oct 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>DeepAgents</category><category>LangChain</category><category>LangGraph</category><category>AI代理</category><category>Python</category><category>多智能体系统</category><category>工具包</category><category>代理框架</category><category>代理运行时</category><category>DeepSeek</category><category>Tavily</category><category>任务规划</category><category>上下文管理</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-10-05-Foundation-Sprint</guid>
    <title>Foundation Sprint：10小时搞定产品战略的谷歌秘籍</title>
    <link>https://apframework.com/blog/essay/2025-10-05-Foundation-Sprint</link>
    <description>来自谷歌风投的 Foundation Sprint 方法论，用短短10小时让核心团队达成战略共识，为产品成功奠定基础。本文详细介绍这一高效的产品规划流程，包含实战案例和具体操作方法。</description>
    <pubDate>Sun, 05 Oct 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>产品管理</category><category>创业方法论</category><category>Design Sprint</category><category>产品战略</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-10-04-development-laws</guid>
    <title>软件开发七大经典定律：从项目管理到系统设计的实践指南</title>
    <link>https://apframework.com/blog/essay/2025-10-04-development-laws</link>
    <description>深入解析软件开发领域最具指导意义的七大经典定律，从项目管理到系统设计的全方位实践指南</description>
    <pubDate>Sat, 04 Oct 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>架构</category><category>管理</category><category>开发</category><category>定律</category><category>软件工程</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-09-12-langgraph-concept-subgraph</guid>
    <title>LangGraph 子图（Subgraph）：概念总结与代码验证</title>
    <link>https://apframework.com/blog/essay/2025-09-12-langgraph-concept-subgraph</link>
    <description>本文总结 LangGraph 子图（Subgraph）能力，并通过两种通信模式（共享状态与不同状态）提供完整可运行的 Python 示例与流式输出，展示子图在父图中的编译与调用方式、输入输出映射以及在多智能体系统中的常见用法，帮助你快速理解并验证该特性。</description>
    <pubDate>Fri, 12 Sep 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>LangGraph</category><category>Subgraph</category><category>LLM</category><category>状态机</category><category>智能体</category><category>Python</category><category>教程</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-09-02-langgraph-concept-mcp</guid>
    <title>LangGraph 集成 FastMCP 2.0 基础教程：两种集成方式的完整实践</title>
    <link>https://apframework.com/blog/essay/2025-09-02-langgraph-concept-mcp</link>
    <description>本文作为基础教程，系统讲解 LangGraph 与 FastMCP 2.0（MCP 2.0） 的两种集成方式：作为客户端在 LangGraph 工作流中调用 MCP 工具，以及将现有 LangGraph Agent 封装为 FastMCP 服务器对外提供服务。内容包含环境准备、示例代码、运行验证与 MCP Inspector 调试，帮助你快速上手并完成端到端打通。</description>
    <pubDate>Tue, 02 Sep 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>LangGraph</category><category>FastMCP</category><category>MCP</category><category>Agent</category><category>Python</category><category>工具集成</category><category>模型上下文协议</category><category>教程</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-08-23-Deep-Research</guid>
    <title>Deep Research 深度解析：基于 AI 的智能研究报告生成平台</title>
    <link>https://apframework.com/blog/essay/2025-08-23-Deep-Research</link>
    <description>Deep Research 是一个基于大语言模型的深度研究报告生成平台。本文通过详细的架构分析和技术实现细节，为开发者提供构建智能研究助手系统的最佳实践参考。</description>
    <pubDate>Sat, 23 Aug 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>Deep Research</category><category>AI</category><category>研究报告</category><category>大语言模型</category><category>Next.js</category><category>Vercel</category><category>多模型架构</category><category>TypeScript</category><category>知识图谱</category><category>RAG</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-08-14-open-deep-research</guid>
    <title>LangChain Open Deep Research：开源深度研究系统</title>
    <link>https://apframework.com/blog/essay/2025-08-14-open-deep-research</link>
    <description>深入解析 LangChain Open Deep Research 开源深度研究代理系统的三阶段架构设计：范围界定、研究执行和报告撰写。详细剖析多代理协作机制、工具集成策略和代码实现，为构建智能研究助手提供完整技术方案和最佳实践指导。</description>
    <pubDate>Thu, 14 Aug 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>LangChain</category><category>Open Deep Research</category><category>Agent</category><category>AI</category><category>深度研究</category><category>多代理系统</category><category>MCP</category><category>架构设计</category><category>Python</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-08-10-Embabel</guid>
    <title>Embabel 简介：新一代企业级智能代理框架介绍</title>
    <link>https://apframework.com/blog/essay/2025-08-10-Embabel</link>
    <description>介绍 Embabel 智能代理框架的核心目标与设计理念。文章详细介绍了这个由 Spring 框架创始人 Rod Johnson 打造的企业级 AI 代理框架，并通过实际代码示例展示如何构建故事创作与评审代理，为 JVM 生态的 AI 应用开发提供介绍。</description>
    <pubDate>Sun, 10 Aug 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>Embabel</category><category>AI代理</category><category>JVM</category><category>Spring</category><category>Kotlin</category><category>Java</category><category>企业级AI</category><category>智能规划</category><category>领域建模</category><category>MCP</category><category>A2A</category><category>生成式AI</category><category>Rod Johnson</category><category>OODA循环</category><category>代理框架</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-07-29-DeerFlow</guid>
    <title>DeerFlow 深度解析：基于 LangGraph 的多智能体研究系统</title>
    <link>https://apframework.com/blog/essay/2025-07-29-DeerFlow</link>
    <description>深入解析 DeerFlow 多智能体研究系统的技术架构与实现细节。文章详细剖析了基于 LangGraph构 建的模块化工作流，从协调器、规划器到研究团队的完整智能体协作机制，涵盖Python现代技术栈、异步处理等核心技术。通过详细流程分析和独立入口函数设计分析，展示了如何构建生产级的AI研究助手系统，为开发者提供多智能体系统设计的最佳实践参考。</description>
    <pubDate>Tue, 29 Jul 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>DeerFlow</category><category>LangGraph</category><category>多智能体系统</category><category>AI架构</category><category>Python</category><category>FastAPI</category><category>状态机</category><category>工作流编排</category><category>RAG</category><category>智能体协作</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-07-13-BAML</guid>
    <title>BAML 原理与实战：AI 工作流与代理的声明式提示语言全解析</title>
    <link>https://apframework.com/blog/essay/2025-07-13-BAML</link>
    <description>全面介绍 BAML（BoundaryML Application Modeling Language）声明式提示语言的设计理念、安装与使用方法，并结合源码与官网文档，深入剖析其类型系统、编译流程与 Python 集成原理，帮助开发者高效构建可靠的 AI 工作流与智能代理。</description>
    <pubDate>Sun, 13 Jul 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>BAML</category><category>AI工作流</category><category>提示工程</category><category>代理</category><category>LLM</category><category>Python</category><category>类型安全</category><category>BoundaryML</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-07-08-12-Factor-Agents</guid>
    <title>12-Factor Agents：构建生产级 LLM 代理的 12 要素原则</title>
    <link>https://apframework.com/blog/essay/2025-07-08-12-Factor-Agents</link>
    <description>本文为 12-Factor Agents 英文原文的翻译整理，并结合个人实践补充了部分笔记和理解。内容涵盖 LLM 代理开发的 12 项核心原则，适合关注 AI 工程、生产级 Agent、AI 应用最佳实践的开发者参考。</description>
    <pubDate>Tue, 08 Jul 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>12-Factor Agents</category><category>LLM应用</category><category>AI代理</category><category>生产级Agent</category><category>AI工程</category><category>最佳实践</category><category>大语言模型</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-07-02-context</guid>
    <title>上下文工程：从提示工程到智能体时代的关键技术</title>
    <link>https://apframework.com/blog/essay/2025-07-02-context</link>
    <description>本文探讨了上下文工程的核心概念与实践方法，阐述了从提示工程到上下文工程的演进过程，详细分析了上下文的组成要素、重要性及其在智能体开发中的关键作用，并通过LangGraph等实例展示了如何有效实现上下文管理。</description>
    <pubDate>Wed, 02 Jul 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>上下文工程</category><category>大语言模型</category><category>LLM</category><category>智能体</category><category>RAG</category><category>提示工程</category><category>记忆系统</category><category>LangGraph</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-06-22-mem0</guid>
    <title>Mem0 深度解析：开源本地 AI 智能体长期记忆系统原理与实战</title>
    <link>https://apframework.com/blog/essay/2025-06-22-mem0</link>
    <description>本文全面解析 Mem0（mem-zero）开源本地 AI 记忆管理系统的技术原理与架构设计，详细介绍如何在本地环境下安全部署 Mem0，实现 AI 智能体的长期记忆与上下文感知。文章涵盖 Mem0 的核心原理、MCP 协议、向量存储与图数据库集成，并通过完整的本地搭建流程、代码实操、LangGraph 与 Dify 集成案例，帮助开发者快速构建具备持久记忆与个性化交互能力的 AI 应用。</description>
    <pubDate>Sun, 22 Jun 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>Mem0</category><category>OpenMemory</category><category>AI记忆</category><category>MCP</category><category>向量存储</category><category>智能体</category><category>Python</category><category>本地部署</category><category>LangGraph</category><category>Dify</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/notes/2025-06-17-Q-Learning</guid>
    <title>Q-Learning 精解：从数学原理到代码实现的深度探索</title>
    <link>https://apframework.com/blog/notes/2025-06-17-Q-Learning</link>
    <description>本文系统剖析Q-Learning算法的核心机制与理论基础，通过冰冻湖面环境的实例演示，从贝尔曼方程到代码实现，为读者构建强化学习的完整知识体系，助力AI学习之旅。</description>
    <pubDate>Tue, 17 Jun 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>强化学习</category><category>Q-Learning</category><category>人工智能</category><category>机器学习</category><category>马尔可夫决策过程</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/notes/2025-06-15-Monte-Carlo-method</guid>
    <title>蒙特卡罗方法：原理解析与Python实现</title>
    <link>https://apframework.com/blog/notes/2025-06-15-Monte-Carlo-method</link>
    <description>本文详细介绍了蒙特卡罗方法的基本原理和应用场景，并通过Python代码实现了π值估算和定积分计算两个经典案例，展示了该方法在实际问题中的应用。</description>
    <pubDate>Sun, 15 Jun 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>强化学习</category><category>蒙特卡罗方法</category><category>Python</category><category>数值计算</category><category>随机模拟</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-06-11-Python-MCP</guid>
    <title>深入浅出 Python MCP SDK: 构建可扩展的分布式服务端与客户端</title>
    <link>https://apframework.com/blog/essay/2025-06-11-Python-MCP</link>
    <description>本文系统化探讨 Model Context Protocol (MCP) Python SDK 的核心原理与实现机制，从 Streamable HTTP 传输协议到异步服务器与客户端开发全过程。通过电灯开关服务的实战案例，详解 ASGI 集成、资源管理、工具调用与会话管理等核心技术，为开发者提供构建高性能分布式应用的实用指南。</description>
    <pubDate>Wed, 11 Jun 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>Python</category><category>MCP</category><category>Streamable HTTP</category><category>异步编程</category><category>服务器开发</category><category>客户端开发</category><category>ASGI</category><category>分布式系统</category><category>实时通信</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-06-08-LangGraph-Tutorials</guid>
    <title>LangGraph实战：从零开始构建智能体应用的状态机编排指南</title>
    <link>https://apframework.com/blog/essay/2025-06-08-LangGraph-Tutorials</link>
    <description>本文全面解析LangGraph状态机框架，从理论到实践，提供详尽代码示例与运行指南。通过循序渐进的实战演练，掌握从基础聊天机器人到高级状态管理的完整开发流程，深度探索如何结合DeepSeek等大模型构建具有持久记忆与工具集成能力的智能应用，助力开发者快速驾驭这一前沿AI编排技术。</description>
    <pubDate>Sun, 08 Jun 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>LangGraph</category><category>LLM</category><category>状态机</category><category>智能体</category><category>DeepSeek</category><category>Python</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-06-07-Python-UV</guid>
    <title>UV：下一代Python包管理工具的全面指南</title>
    <link>https://apframework.com/blog/essay/2025-06-07-Python-UV</link>
    <description>深入介绍了 UV 作为新一代Python包管理工具的核心特性、安装配置、基本概念及实际应用案例，帮助开发者提升依赖管理效率，解决传统工具的痛点问题，加速Python项目开发流程。</description>
    <pubDate>Sat, 07 Jun 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>Python</category><category>UV</category><category>包管理</category><category>虚拟环境</category><category>依赖管理</category><category>Python工具链</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-04-26-Python-Poetry</guid>
    <title>Poetry：现代Python依赖管理工具的实践指南</title>
    <link>https://apframework.com/blog/essay/2025-04-26-Python-Poetry</link>
    <description>全面剖析Poetry作为Python项目依赖管理工具的核心功能、最佳实践及与Conda的协同工作策略，帮助开发者构建稳定可靠的项目环境，解决依赖冲突问题，提升开发效率。</description>
    <pubDate>Sat, 26 Apr 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>Python</category><category>Poetry</category><category>依赖管理</category><category>虚拟环境</category><category>包管理</category><category>pyproject.toml</category><category>Python工具链</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-04-17-Java-MCP</guid>
    <title>Java MCP SDK实战指南: 构建大模型与外部工具的智能连接</title>
    <link>https://apframework.com/blog/essay/2025-04-17-Java-MCP</link>
    <description>本文全面解析Model Context Protocol (MCP) Java SDK的核心原理与实现方法，详细讲解如何使用Spring MCP构建连接大型语言模型与外部工具的桥梁，并通过实现天气服务API的实战案例，帮助开发者快速掌握MCP的实际应用与最佳实践。</description>
    <pubDate>Fri, 18 Apr 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>Java</category><category>MCP</category><category>Spring Boot</category><category>AI集成</category><category>大语言模型</category><category>开发工具</category><category>LLM</category><category>Spring AI</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-04-09-18-RAG-Techniques</guid>
    <title>18种RAG技术实战对比：从简单RAG到自适应RAG全面解析</title>
    <link>https://apframework.com/blog/essay/2025-04-09-18-RAG-Techniques</link>
    <description>详细解析18种RAG技术实现与对比，从基础RAG到高级自适应RAG，展示不同维度的实现思路。全文使用基础库而非框架构建，附带完整代码示例，是理解RAG技术原理与实践的理想入门指南。</description>
    <pubDate>Wed, 09 Apr 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>RAG技术</category><category>HyDE</category><category>RSE</category><category>自我RAG</category><category>自适应RAG</category><category>重排序器</category><category>图RAG</category><category>层次RAG</category><category>CRAG</category><category>LLM应用</category><category>检索增强生成</category><category>AI开发</category><category>大语言模型</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-04-06-Dify-MCP-ClickHouse</guid>
    <title>Dify 中使用 MCP 协议访问 ClickHouse 搭建完整教程</title>
    <link>https://apframework.com/blog/essay/2025-04-06-Dify-MCP-ClickHouse</link>
    <description>详细记录如何在 Dify 大语言模型应用开发平台中，通过 MCP 协议及 MCP Proxy 连接 ClickHouse 数据库，实现数据分析与可视化的完整教程。包含环境配置、服务部署和实际应用案例。</description>
    <pubDate>Sun, 06 Apr 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>Dify</category><category>MCP</category><category>ClickHouse</category><category>ECharts</category><category>RAG</category><category>数据可视化</category><category>AI应用开发</category><category>大语言模型</category><category>LLM工具集成</category><category>数据分析</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-04-02-Facebook-Prophet</guid>
    <title>Facebook Prophet 时间序列预测：商品价格分析与实战指南</title>
    <link>https://apframework.com/blog/essay/2025-04-02-Facebook-Prophet</link>
    <description>全面解析Meta(原Facebook)开源的Prophet时间序列预测工具，从理论原理到实际应用。学习如何利用Prophet的加法模型进行商品价格预测，掌握数据预处理、模型配置与结果分析的完整流程，附带Python代码示例与详细解释。</description>
    <pubDate>Wed, 02 Apr 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>Prophet</category><category>Facebook</category><category>时间序列预测</category><category>数据分析</category><category>机器学习</category><category>价格预测</category><category>加法模型</category><category>趋势分析</category><category>季节性分析</category><category>Python</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2025-02-16-Google-Agent</guid>
    <title>Google Agent白皮书解析：模型、工具与编排层全面指南</title>
    <link>https://apframework.com/blog/essay/2025-02-16-Google-Agent</link>
    <description>深入解析Google Agent白皮书核心内容，详细介绍智能代理的架构、组件和工作原理，包括模型、工具、编排层以及ReAct、CoT和ToT等推理框架的应用与区别。</description>
    <pubDate>Sun, 16 Feb 2025 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>AI</category><category>Agent</category><category>LLM</category><category>Google</category><category>人工智能</category><category>大语言模型</category><category>智能代理</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/notes/2024-12-24-Generative-AI-on-AWS</guid>
    <title>生成式AI入门与AWS实战 阅读笔记</title>
    <link>https://apframework.com/blog/notes/2024-12-24-Generative-AI-on-AWS</link>
    <description>《生成式AI入门与AWS实战》（Generative AI on AWS） 的阅读笔记</description>
    <pubDate>Sun, 22 Dec 2024 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>AI</category><category>Generative AI</category><category>生成式AI</category><category>LLM</category><category>乱翻书</category><category>读书笔记</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2024-12-15-Claude-MCP</guid>
    <title>Claude MCP 使用及搭建</title>
    <link>https://apframework.com/blog/essay/2024-12-15-Claude-MCP</link>
    <description>Model Context Protocol (MCP)</description>
    <pubDate>Sun, 15 Dec 2024 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>AI</category><category>MCP</category><category>Prompt</category><category>LLM</category><category>随手记</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2024-12-08-COLA-Light</guid>
    <title>COLA Light 5.0实践指南：轻量级DDD分层架构详解</title>
    <link>https://apframework.com/blog/essay/2024-12-08-COLA-Light</link>
    <description>详细解析COLA Light 5.0轻量级分层架构的实践应用，包含完整的项目创建流程、分层架构原理及实战经验，帮助开发团队快速构建符合DDD理念的应用架构</description>
    <pubDate>Sun, 08 Dec 2024 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>COLA</category><category>DDD</category><category>分层架构</category><category>领域驱动设计</category><category>微服务</category><category>Java架构</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/essay/2024-11-16-ESLint-theory</guid>
    <title>ESLint 概念及原理探索：JavaScript 代码质量工具</title>
    <link>https://apframework.com/blog/essay/2024-11-16-ESLint-theory</link>
    <description>ESLint 是前端开发中不可或缺的代码质量工具，本文深入探讨其概念、工作原理及实现机制，帮助开发者更好地理解和使用这一强大的 JavaScript 检查器。</description>
    <pubDate>Sat, 16 Nov 2024 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>ESLint</category><category>JavaScript</category><category>前端工具</category><category>代码质量</category><category>静态分析</category>
  </item>

  <item>
    <guid>https://apframework.com/blog/opensource/2018-1-9-mweb-github-jekyll</guid>
    <title>GitHub Pages 搭建笔记</title>
    <link>https://apframework.com/blog/opensource/2018-1-9-mweb-github-jekyll</link>
    <description>Jekyll、Github.io、七牛云 构建个人博客</description>
    <pubDate>Mon, 08 Jan 2018 00:00:00 GMT</pubDate>
    <author>huangshoukai@yeah.net (Shoukai Huang)</author>
    <category>环境搭建</category><category>GitHub Pages</category><category>Blog</category>
  </item>

    </channel>
  </rss>
