🐦 Twitter Post Details

Viewing enriched Twitter post

@dair_ai

Banger paper from Microsoft. It's on agent reliability in real business workflows. (bookmark it) Thinkingbox is a sandbox with isolated MCP-compatible tool sessions, plus a benchmark of 507 policy-conditioned workflows across retail, hospitality, auto insurance, neobank IT, and consulting support. Every attempt is graded on the backend state the agent leaves behind. Executable checks accept valid trajectories and reject wrong, missing, or extra effects, so collateral damage counts against you. The strongest model reaches 65.36% pass@1 and 25.25% pass^20. Many failed trials terminate cleanly with valid state-changing tool calls. Watching the response or the tool call tells you very little about whether the task actually completed. Paper: https://t.co/EkMhBafbvI Track more trending AI papers in our academy: https://t.co/LRnpZN7L4c

Media 1
Media 2
View on Twitter

📊 Media Metadata

{
  "media": [
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2091209556122808532/media_0.jpg",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2091209556122808532/media_0.jpg",
      "type": "photo",
      "filename": "media_0.jpg"
    },
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2091209556122808532/media_2.jpg",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2091209556122808532/media_2.jpg",
      "type": "photo",
      "filename": "media_2.jpg"
    }
  ],
  "processed_at": "2026-08-22T18:02:12.578007",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2091209556122808532",
  "url": "https://x.com/dair_ai/status/2091209556122808532",
  "twitterUrl": "https://twitter.com/dair_ai/status/2091209556122808532",
  "text": "Banger paper from Microsoft.\n\nIt's on agent reliability in real business workflows.\n\n(bookmark it)\n\nThinkingbox is a sandbox with isolated MCP-compatible tool sessions, plus a benchmark of 507 policy-conditioned workflows across retail, hospitality, auto insurance, neobank IT, and consulting support.\n\nEvery attempt is graded on the backend state the agent leaves behind. Executable checks accept valid trajectories and reject wrong, missing, or extra effects, so collateral damage counts against you.\n\nThe strongest model reaches 65.36% pass@1 and 25.25% pass^20.\n\nMany failed trials terminate cleanly with valid state-changing tool calls. Watching the response or the tool call tells you very little about whether the task actually completed.\n\nPaper: https://t.co/EkMhBafbvI\n\nTrack more trending AI papers in our academy: https://t.co/LRnpZN7L4c",
  "source": "Twitter for iPhone",
  "retweetCount": 4,
  "replyCount": 9,
  "likeCount": 28,
  "quoteCount": 0,
  "viewCount": 2939,
  "createdAt": "Sat Aug 22 17:03:05 +0000 2026",
  "lang": "en",
  "bookmarkCount": 34,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2091209556122808532",
  "displayTextRange": [
    0,
    277
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "dair_ai",
    "url": "https://x.com/dair_ai",
    "twitterUrl": "https://twitter.com/dair_ai",
    "id": "889050642903293953",
    "name": "DAIR.AI",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/1643277398522187778/31dedbLo_normal.jpg",
    "coverPicture": "https://pbs.twimg.com/profile_banners/889050642903293953/1773242460",
    "description": "",
    "location": "",
    "followers": 129859,
    "following": 1,
    "status": "",
    "canDm": true,
    "canMediaTag": true,
    "createdAt": "Sun Jul 23 09:12:45 +0000 2017",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 5069,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 321,
    "statusesCount": 3614,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "2091209556122808532"
    ],
    "profile_bio": {
      "description": "Democratizing AI research, education, and technologies. Learn about AI Agents for FREE at https://t.co/HHXg8rryu4",
      "entities": {
        "description": {
          "urls": [
            {
              "display_url": "academy.dair.ai/courses/elemen…",
              "expanded_url": "https://academy.dair.ai/courses/elements-of-ai-agents",
              "indices": [
                90,
                113
              ],
              "url": "https://t.co/HHXg8rryu4"
            }
          ]
        },
        "url": {
          "urls": [
            {
              "display_url": "dair.ai",
              "expanded_url": "https://www.dair.ai/",
              "indices": [
                0,
                23
              ],
              "url": "https://t.co/lkqPZtMU5s"
            }
          ]
        }
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "display_url": "pic.x.com/lNZWCMSawE",
        "expanded_url": "https://x.com/dair_ai/status/2091209556122808532/photo/1",
        "ext_media_availability": {
          "status": "Available"
        },
        "features": {
          "large": {
            "faces": []
          },
          "medium": {
            "faces": []
          },
          "orig": {
            "faces": []
          },
          "small": {
            "faces": []
          }
        },
        "id_str": "2091209543162486785",
        "indices": [
          278,
          301
        ],
        "media_key": "3_2091209543162486785",
        "media_results": {
          "result": {
            "media_key": "3_2091209543162486785"
          }
        },
        "media_url_https": "https://pbs.twimg.com/media/HQV3__MbYAE2htR.png",
        "original_info": {
          "focus_rects": [
            {
              "h": 923,
              "w": 1648,
              "x": 0,
              "y": 0
            },
            {
              "h": 1648,
              "w": 1648,
              "x": 0,
              "y": 0
            },
            {
              "h": 1862,
              "w": 1633,
              "x": 0,
              "y": 0
            },
            {
              "h": 1862,
              "w": 931,
              "x": 326,
              "y": 0
            },
            {
              "h": 1862,
              "w": 1648,
              "x": 0,
              "y": 0
            }
          ],
          "height": 1862,
          "width": 1648
        },
        "sizes": {
          "large": {
            "h": 1862,
            "resize": "fit",
            "w": 1648
          },
          "medium": {
            "h": 1200,
            "resize": "fit",
            "w": 1062
          },
          "small": {
            "h": 680,
            "resize": "fit",
            "w": 602
          },
          "thumb": {
            "h": 150,
            "resize": "crop",
            "w": 150
          }
        },
        "type": "photo",
        "url": "https://t.co/lNZWCMSawE"
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "smarttags": [],
    "symbols": [],
    "urls": [
      {
        "display_url": "arxiv.org/abs/2608.19741",
        "expanded_url": "https://arxiv.org/abs/2608.19741",
        "indices": [
          754,
          777
        ],
        "url": "https://t.co/EkMhBafbvI"
      },
      {
        "display_url": "academy.dair.ai",
        "expanded_url": "https://academy.dair.ai/",
        "indices": [
          825,
          848
        ],
        "url": "https://t.co/LRnpZN7L4c"
      }
    ],
    "user_mentions": []
  },
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "article": null
}