🐦 Twitter Post Details

Viewing enriched Twitter post

@0xViviennn

在用 Hermes Agent 的时候,一直感觉跨 session 的中文搜索不太好用。Agent明明说过的话,session_search 就是搜不到,像完全失忆了一样.... 比如昨天搞了一天的A2A通信,但今天再搜"和其他Agent的聊天记录"的时候,返回 0 条结果。但数据确实在数据库里——用 LIKE子串匹配能搜出 20 多条 然后我让 Claude Code 帮忙查了一下,原因是 SQLite FTS5的默认分词器。英文有空格,天然能按词分 中文没有空格,FTS5把每个汉字当成独立 token。搜"记忆断裂"实际执行的是 记 & 忆 & 断 & 裂——四个独立的字分别匹配,丢失了词的连续性,搜不到原本存在的内容 日文、韩文也一样,都没有空格分隔,都会中招,但英文用户完全不受影响。 目前的解决方案是 FTS5 搜不到结果时,检测到查询包含 CJK 字符就自动降级到 LIKE子串匹配。比 FTS5 慢一点(全表扫描),但对 agent 的数据量完全够用。 已提 PR: https://t.co/QgT9roKg6P @NousResearch @Teknium

Media 1
Media 2

📊 Media Metadata

{
  "media": [
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2045082042162139547/media_0.jpg",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2045082042162139547/media_0.jpg",
      "type": "photo",
      "filename": "media_0.jpg"
    },
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2045082042162139547/media_1.jpg",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2045082042162139547/media_1.jpg",
      "type": "photo",
      "filename": "media_1.jpg"
    }
  ],
  "processed_at": "2026-04-18T09:13:26.468057",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2045082042162139547",
  "url": "https://x.com/0xViviennn/status/2045082042162139547",
  "twitterUrl": "https://twitter.com/0xViviennn/status/2045082042162139547",
  "text": "在用 Hermes Agent 的时候,一直感觉跨 session 的中文搜索不太好用。Agent明明说过的话,session_search 就是搜不到,像完全失忆了一样....\n\n比如昨天搞了一天的A2A通信,但今天再搜\"和其他Agent的聊天记录\"的时候,返回 0 条结果。但数据确实在数据库里——用 LIKE子串匹配能搜出 20 多条\n\n然后我让 Claude Code 帮忙查了一下,原因是 SQLite FTS5的默认分词器。英文有空格,天然能按词分\n\n中文没有空格,FTS5把每个汉字当成独立 token。搜\"记忆断裂\"实际执行的是 记 & 忆 & 断 & 裂——四个独立的字分别匹配,丢失了词的连续性,搜不到原本存在的内容\n\n日文、韩文也一样,都没有空格分隔,都会中招,但英文用户完全不受影响。\n\n目前的解决方案是 FTS5 搜不到结果时,检测到查询包含 CJK 字符就自动降级到 LIKE子串匹配。比 FTS5 慢一点(全表扫描),但对 agent 的数据量完全够用。\n\n已提 PR:\nhttps://t.co/QgT9roKg6P\n\n@NousResearch @Teknium",
  "source": "Twitter for iPhone",
  "retweetCount": 2,
  "replyCount": 17,
  "likeCount": 78,
  "quoteCount": 2,
  "viewCount": 9732,
  "createdAt": "Fri Apr 17 10:08:49 +0000 2026",
  "lang": "zh",
  "bookmarkCount": 56,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2045082042162139547",
  "displayTextRange": [
    0,
    176
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "0xViviennn",
    "url": "https://x.com/0xViviennn",
    "twitterUrl": "https://twitter.com/0xViviennn",
    "id": "1888243896368644096",
    "name": "Viviennn",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/1896938905884299265/x3BzkUic_normal.jpg",
    "coverPicture": "https://pbs.twimg.com/profile_banners/1888243896368644096/1741104307",
    "description": "",
    "location": "",
    "followers": 685,
    "following": 2007,
    "status": "",
    "canDm": true,
    "canMediaTag": true,
    "createdAt": "Sat Feb 08 15:10:13 +0000 2025",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 4856,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 121,
    "statusesCount": 762,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "2038657899963281725"
    ],
    "profile_bio": {
      "description": "✷ 𝚜𝚌𝚑𝚒𝚣𝚘𝚙𝚘𝚜𝚝𝚎𝚛 ✷",
      "entities": {
        "description": {}
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "allow_download_status": {
          "allow_download": true
        },
        "display_url": "pic.twitter.com/yIaBYvTQhZ",
        "expanded_url": "https://twitter.com/0xViviennn/status/2045082042162139547/photo/1",
        "ext_media_availability": {
          "status": "Available"
        },
        "features": {
          "large": {
            "faces": [
              {
                "h": 47,
                "w": 47,
                "x": 101,
                "y": 9
              }
            ]
          },
          "orig": {
            "faces": [
              {
                "h": 47,
                "w": 47,
                "x": 101,
                "y": 9
              }
            ]
          }
        },
        "id_str": "2045081960310300672",
        "indices": [
          177,
          200
        ],
        "media_key": "3_2045081960310300672",
        "media_results": {
          "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARxhlzUk2uAACgACHGGXSDOa0ZsAAA==",
          "result": {
            "__typename": "ApiMedia",
            "id": "QXBpTWVkaWE6DAABCgABHGGXNSTa4AAKAAIcYZdIM5rRmwAA",
            "media_key": "3_2045081960310300672"
          }
        },
        "media_url_https": "https://pbs.twimg.com/media/HGGXNSTa4AAYh36.jpg",
        "original_info": {
          "focus_rects": [
            {
              "h": 608,
              "w": 1086,
              "x": 0,
              "y": 0
            },
            {
              "h": 730,
              "w": 730,
              "x": 0,
              "y": 0
            },
            {
              "h": 730,
              "w": 640,
              "x": 0,
              "y": 0
            },
            {
              "h": 730,
              "w": 365,
              "x": 62,
              "y": 0
            },
            {
              "h": 730,
              "w": 1086,
              "x": 0,
              "y": 0
            }
          ],
          "height": 730,
          "width": 1086
        },
        "sizes": {
          "large": {
            "h": 730,
            "w": 1086
          }
        },
        "type": "photo",
        "url": "https://t.co/yIaBYvTQhZ"
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "urls": [
      {
        "display_url": "github.com/NousResearch/h…",
        "expanded_url": "http://github.com/NousResearch/hermes-agent/pull/11517",
        "indices": [
          451,
          474
        ],
        "url": "https://t.co/QgT9roKg6P"
      }
    ],
    "user_mentions": [
      {
        "id_str": "1318419526132862976",
        "indices": [
          476,
          489
        ],
        "name": "Nous Research",
        "screen_name": "NousResearch"
      },
      {
        "id_str": "1365020011123773442",
        "indices": [
          490,
          498
        ],
        "name": "Teknium (e/λ)",
        "screen_name": "Teknium"
      }
    ]
  },
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "communityInfo": null,
  "article": null
}