Hugging Face Daily Papers · May 13, 2026 · 3 min read

MCP-Cosmos: World Model-Augmented Agents for Complex Task Execution in MCP Environments

Mirrored from Hugging Face Daily Papers for archival readability. Support the source by reading on the original site.

Like Read original ↗

World Model 4 MCP</p>\n","updatedAt":"2026-05-13T03:16:37.002Z","author":{"_id":"64c47f731d44fc06afc80953","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/UT2mHX2WuCm5Ws4rGKyCB.png","fullname":"Dhaval Patel","name":"DhavalPatel","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":7,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.38307052850723267},"editors":["DhavalPatel"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/UT2mHX2WuCm5Ws4rGKyCB.png"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2605.09131","authors":[{"_id":"6a03ece286b054ce2fa40e82","name":"Giridhar Ganapavarapu","hidden":false},{"_id":"6a03ece286b054ce2fa40e83","name":"Dhaval Patel","hidden":false}],"publishedAt":"2026-05-09T00:00:00.000Z","submittedOnDailyAt":"2026-05-13T00:00:00.000Z","title":"MCP-Cosmos: World Model-Augmented Agents for Complex Task Execution in MCP Environments","submittedOnDailyBy":{"_id":"64c47f731d44fc06afc80953","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/UT2mHX2WuCm5Ws4rGKyCB.png","isPro":false,"fullname":"Dhaval Patel","user":"DhavalPatel","type":"user","name":"DhavalPatel"},"summary":"The Model Context Protocol (MCP) has unified the interface between Large Language Models (LLMs) and external tools, yet a fundamental gap remains in how agents conceptualize the environments within which they operate. Current paradigms are bifurcated: Task-level planning often ignores execution-time dynamics, while reactive execution lacks long-horizon foresight. We present MCP-Cosmos, a framework that infuses generative World Models (WM) into the MCP ecosystem to enable predictive task automation. By unifying three disparate technologies, namely MCP, World Model, and Agent, we demonstrate that a \"Bring Your Own World Model\" (BYOWM) strategy allows agents to simulate state transitions and refine plans in a latent space before execution. We conducted experiments using two strategies, namely ReAct and SPIRAL with 2 planning models and 3 representative world models over 20+ MCP-Bench tasks. We observed improvements in Agent's environment interaction KPI such as tool success rate and tool parameter accuracy. The framework also offers new metrics such as Execution Quality to generate new insights about the effectiveness of world models compared to baseline.","upvotes":22,"discussionId":"6a03ece286b054ce2fa40e84","ai_summary":"MCP-Cosmos integrates generative World Models into the Model Context Protocol ecosystem to enhance agent planning and execution through predictive simulation in latent space.","ai_keywords":["Model Context Protocol","Large Language Models","generative World Models","agent planning","task automation","latent space","ReAct","SPIRAL","tool success rate","tool parameter accuracy","Execution Quality","BYOWM"],"organization":{"_id":"6760ab6c5c9a8ea8370ab95b","name":"ibm-research","fullname":"IBM Research","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/637bfdf60dc13843b468ac20/npxapKcW-cXX3J2JBl2vY.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"64c47f731d44fc06afc80953","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/UT2mHX2WuCm5Ws4rGKyCB.png","isPro":false,"fullname":"Dhaval Patel","user":"DhavalPatel","type":"user"},{"_id":"64dfbcb18e2084e1d7b51b46","avatarUrl":"/avatars/fafe30beea2d7e8eec3f3ba985c582f7.svg","isPro":false,"fullname":"Kushal Raj Bhandari","user":"KBhandari11","type":"user"},{"_id":"66739484a660dfbb2643eb3d","avatarUrl":"/avatars/c9a6c6a1e295e448dacfccb1a2860e8a.svg","isPro":false,"fullname":"Leo Y","user":"LeoYML","type":"user"},{"_id":"677629b4931b3b2eb8d5820a","avatarUrl":"/avatars/ed9723b5fd015bb65e8dceb0a6d86ff9.svg","isPro":false,"fullname":"Ankita Naik","user":"ankita-naik","type":"user"},{"_id":"662436602d61edba3d27e263","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/IpikJRhMDkaaAvVMiaVj6.png","isPro":false,"fullname":"Chathurangi Shyalika","user":"ChathurangiShyalika","type":"user"},{"_id":"64d1140073dc458c16489ca0","avatarUrl":"/avatars/c1ad9d970a293fc6d8e2acfdb61c352b.svg","isPro":false,"fullname":"Rohan Arora","user":"rohan-arora","type":"user"},{"_id":"620783f24e28382272337ba4","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/620783f24e28382272337ba4/zkUveQPNiDfYjgGhuFErj.jpeg","isPro":false,"fullname":"GuoLiangTang","user":"Tommy930","type":"user"},{"_id":"658e7d80e7b4143ec87fa01e","avatarUrl":"/avatars/fa5834d4282fde8d5affa24e17494590.svg","isPro":false,"fullname":"Lilian Ngweta","user":"lilianngweta","type":"user"},{"_id":"6836e2b6f307012b46561f99","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/qAs7KReYW3yHnhgH_nzv4.png","isPro":false,"fullname":"Vinamra Baghel","user":"vinamra-baghel","type":"user"},{"_id":"6407e5294edf9f5c4fd32228","avatarUrl":"/avatars/8e2d55460e9fe9c426eb552baf4b2cb0.svg","isPro":false,"fullname":"Stoney Kang","user":"sikang99","type":"user"},{"_id":"6141a88b3a0ec78603c9e784","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6141a88b3a0ec78603c9e784/DJsxSmWV39M33JFheLobC.jpeg","isPro":true,"fullname":"merve","user":"merve","type":"user"},{"_id":"6826c3d7a677fa26bd486db6","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/vy5ixwWBqZAnGxVVtKx9m.png","isPro":false,"fullname":"James Rayfield","user":"jtrayfield","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"6760ab6c5c9a8ea8370ab95b","name":"ibm-research","fullname":"IBM Research","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/637bfdf60dc13843b468ac20/npxapKcW-cXX3J2JBl2vY.png"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2605/2605.09131.md"}">

Papers

arxiv:2605.09131

MCP-Cosmos: World Model-Augmented Agents for Complex Task Execution in MCP Environments

Published on May 9

· Submitted by

Dhaval Patel on May 13

IBM Research

Upvote

Authors:

Abstract

MCP-Cosmos integrates generative World Models into the Model Context Protocol ecosystem to enhance agent planning and execution through predictive simulation in latent space.

AI-generated summary

The Model Context Protocol (MCP) has unified the interface between Large Language Models (LLMs) and external tools, yet a fundamental gap remains in how agents conceptualize the environments within which they operate. Current paradigms are bifurcated: Task-level planning often ignores execution-time dynamics, while reactive execution lacks long-horizon foresight. We present MCP-Cosmos, a framework that infuses generative World Models (WM) into the MCP ecosystem to enable predictive task automation. By unifying three disparate technologies, namely MCP, World Model, and Agent, we demonstrate that a "Bring Your Own World Model" (BYOWM) strategy allows agents to simulate state transitions and refine plans in a latent space before execution. We conducted experiments using two strategies, namely ReAct and SPIRAL with 2 planning models and 3 representative world models over 20+ MCP-Bench tasks. We observed improvements in Agent's environment interaction KPI such as tool success rate and tool parameter accuracy. The framework also offers new metrics such as Execution Quality to generate new insights about the effectiveness of world models compared to baseline.