
  <rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
    <channel>
      <title>Pony&#39;s Blog</title>
      <link>https://ponyma.io/blog</link>
      <description>在螺旋中上升</description>
      <language>zh-cn</language>
      <managingEditor>m_mone@163.com (Pony Ma)</managingEditor>
      <webMaster>m_mone@163.com (Pony Ma)</webMaster>
      <lastBuildDate>Tue, 26 May 2026 00:00:00 GMT</lastBuildDate>
      <atom:link href="https://ponyma.io/tags/mcp/feed.xml" rel="self" type="application/rss+xml"/>
      
  <item>
    <guid>https://ponyma.io/blog/learn-ai/个人-ai-记忆层-会不会成为下一个-mcp-控制面</guid>
    <title>个人 AI 记忆层，会不会成为下一个 MCP 控制面？</title>
    <link>https://ponyma.io/blog/learn-ai/个人-ai-记忆层-会不会成为下一个-mcp-控制面</link>
    <description>piia-engram、Deja Vu 和 TencentDB Agent Memory 都在做同一件事：把 Agent 的记忆从单个工具里拿出来。这个方向值得看，但还没到下结论的时候。</description>
    <pubDate>Tue, 26 May 2026 00:00:00 GMT</pubDate>
    <author>m_mone@163.com (Pony Ma)</author>
    <category>AI</category><category>Agent</category><category>MCP</category><category>开源</category><category>观点</category>
  </item>

    </channel>
  </rss>
