🐦 Twitter Post Details

Viewing enriched Twitter post

@omarsar0

Great paper on managing agent skills. Skill libraries keep growing, and picking the right skills has become a bottleneck for coding agents. The defaults are to expose the agent to the whole skill collection, or retrieve skills with embeddings and rerankers. Both treat the choice as independent picks. SkillComposer treats composition as one joint decision over which skills, how many, and in what order. A constrained autoregressive decoder over skill identifiers produces the full plan in a single pass, so dependencies between successive skills fall out naturally. On SkillsBench with GPT-5.2-Codex and Gemini-3-Pro-Preview, it lifts pass rate by +23.1 and +18.2pp over no-skill, beats top-3 retrieval, and matches the gold-skill upper bound at lower prompt-token cost. Paper: https://t.co/ovbQf07Mmk Learn to build effective AI agents in our academy: https://t.co/1e8RZKs4uX

Media 1
View on Twitter

📊 Media Metadata

{
  "media": [
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2072430551446032847/media_0.jpg",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2072430551446032847/media_0.jpg",
      "type": "photo",
      "filename": "media_0.jpg"
    }
  ],
  "processed_at": "2026-07-01T22:01:05.603341",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2072430551446032847",
  "url": "https://x.com/omarsar0/status/2072430551446032847",
  "twitterUrl": "https://twitter.com/omarsar0/status/2072430551446032847",
  "text": "Great paper on managing agent skills.\n\nSkill libraries keep growing, and picking the right skills has become a bottleneck for coding agents.\n\nThe defaults are to expose the agent to the whole skill collection, or retrieve skills with embeddings and rerankers. Both treat the choice as independent picks.\n\nSkillComposer treats composition as one joint decision over which skills, how many, and in what order. A constrained autoregressive decoder over skill identifiers produces the full plan in a single pass, so dependencies between successive skills fall out naturally.\n\nOn SkillsBench with GPT-5.2-Codex and Gemini-3-Pro-Preview, it lifts pass rate by +23.1 and +18.2pp over no-skill, beats top-3 retrieval, and matches the gold-skill upper bound at lower prompt-token cost.\n\nPaper: https://t.co/ovbQf07Mmk\n\nLearn to build effective AI agents in our academy: https://t.co/1e8RZKs4uX",
  "source": "Twitter for iPhone",
  "retweetCount": 5,
  "replyCount": 3,
  "likeCount": 23,
  "quoteCount": 0,
  "viewCount": 2586,
  "createdAt": "Wed Jul 01 21:22:02 +0000 2026",
  "lang": "en",
  "bookmarkCount": 30,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2072430551446032847",
  "displayTextRange": [
    0,
    274
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "omarsar0",
    "url": "https://x.com/omarsar0",
    "twitterUrl": "https://twitter.com/omarsar0",
    "id": "3448284313",
    "name": "elvis",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/939313677647282181/vZjFWtAn_normal.jpg",
    "coverPicture": "https://pbs.twimg.com/profile_banners/3448284313/1565974901",
    "description": "",
    "location": "DAIR.AI Academy",
    "followers": 309440,
    "following": 887,
    "status": "",
    "canDm": true,
    "canMediaTag": true,
    "createdAt": "Fri Sep 04 12:59:26 +0000 2015",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 37172,
    "hasCustomTimelines": true,
    "isTranslator": true,
    "mediaCount": 4759,
    "statusesCount": 18543,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "2072400978079261041"
    ],
    "profile_bio": {
      "description": "Building self-improving AI @dair_ai • Prev: Meta AI | PhD • Learn about AI Agents for FREE here: https://t.co/P5SA9u54xO",
      "entities": {
        "description": {
          "urls": [
            {
              "display_url": "academy.dair.ai/courses/elemen…",
              "expanded_url": "https://academy.dair.ai/courses/elements-of-ai-agents",
              "indices": [
                97,
                120
              ],
              "url": "https://t.co/P5SA9u54xO"
            }
          ],
          "user_mentions": [
            {
              "id_str": "",
              "indices": [
                27,
                35
              ],
              "name": "",
              "screen_name": "dair_ai"
            }
          ]
        },
        "url": {
          "urls": [
            {
              "display_url": "dair.ai",
              "expanded_url": "https://www.dair.ai/",
              "indices": [
                0,
                23
              ],
              "url": "https://t.co/XQto5ypkSM"
            }
          ]
        }
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "display_url": "pic.twitter.com/92AY6y8xiy",
        "expanded_url": "https://twitter.com/omarsar0/status/2072430551446032847/photo/1",
        "ext_master_playlist_only": [],
        "ext_media_availability": {
          "status": "Available"
        },
        "ext_playlists": [],
        "features": {
          "large": {
            "faces": [
              {
                "h": 251,
                "w": 251,
                "x": 101,
                "y": 467
              }
            ]
          },
          "orig": {
            "faces": [
              {
                "h": 251,
                "w": 251,
                "x": 101,
                "y": 467
              }
            ]
          }
        },
        "id_str": "2072430547553689600",
        "indices": [
          275,
          298
        ],
        "media_key": "3_2072430547553689600",
        "media_results": {
          "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARzCwJpQmyAACgACHMLAmzibkc8AAA==",
          "result": {
            "__typename": "ApiMedia",
            "id": "QXBpTWVkaWE6DAABCgABHMLAmlCbIAAKAAIcwsCbOJuRzwAA",
            "media_key": "3_2072430547553689600"
          }
        },
        "media_url_https": "https://pbs.twimg.com/media/HMLAmlCbIAA__al.jpg",
        "original_info": {
          "focus_rects": [
            {
              "h": 893,
              "w": 1594,
              "x": 0,
              "y": 0
            },
            {
              "h": 1594,
              "w": 1594,
              "x": 0,
              "y": 0
            },
            {
              "h": 1817,
              "w": 1594,
              "x": 0,
              "y": 0
            },
            {
              "h": 1828,
              "w": 914,
              "x": 340,
              "y": 0
            },
            {
              "h": 1828,
              "w": 1594,
              "x": 0,
              "y": 0
            }
          ],
          "height": 1828,
          "width": 1594
        },
        "sizes": {
          "large": {
            "h": 1828,
            "w": 1594
          }
        },
        "type": "photo",
        "url": "https://t.co/92AY6y8xiy"
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "urls": [
      {
        "display_url": "arxiv.org/abs/2606.32025",
        "expanded_url": "https://arxiv.org/abs/2606.32025",
        "indices": [
          785,
          808
        ],
        "url": "https://t.co/ovbQf07Mmk"
      },
      {
        "display_url": "academy.dair.ai",
        "expanded_url": "https://academy.dair.ai/",
        "indices": [
          861,
          884
        ],
        "url": "https://t.co/1e8RZKs4uX"
      }
    ],
    "user_mentions": []
  },
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "communityInfo": null,
  "article": null
}