{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:TSUYRMZ6GREWAAAC3XNWVHZVTX","short_pith_number":"pith:TSUYRMZ6","canonical_record":{"source":{"id":"2508.14704","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-08-20T13:28:58Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"ee8a8459a50232d35936b97f871d943d5031469396bbf84a27446d89b63f5856","abstract_canon_sha256":"c5d52610af57d6d96de96d3cf828c17fcdc149f2467f93d986d01688edd0a24e"},"schema_version":"1.0"},"canonical_sha256":"9ca988b33e3449600002dddb6a9f359de09745712a622b6d0c96cadfdcb580f6","source":{"kind":"arxiv","id":"2508.14704","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.14704","created_at":"2026-07-05T11:56:37Z"},{"alias_kind":"arxiv_version","alias_value":"2508.14704v1","created_at":"2026-07-05T11:56:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.14704","created_at":"2026-07-05T11:56:37Z"},{"alias_kind":"pith_short_12","alias_value":"TSUYRMZ6GREW","created_at":"2026-07-05T11:56:37Z"},{"alias_kind":"pith_short_16","alias_value":"TSUYRMZ6GREWAAAC","created_at":"2026-07-05T11:56:37Z"},{"alias_kind":"pith_short_8","alias_value":"TSUYRMZ6","created_at":"2026-07-05T11:56:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:TSUYRMZ6GREWAAAC3XNWVHZVTX","target":"record","payload":{"canonical_record":{"source":{"id":"2508.14704","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-08-20T13:28:58Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"ee8a8459a50232d35936b97f871d943d5031469396bbf84a27446d89b63f5856","abstract_canon_sha256":"c5d52610af57d6d96de96d3cf828c17fcdc149f2467f93d986d01688edd0a24e"},"schema_version":"1.0"},"canonical_sha256":"9ca988b33e3449600002dddb6a9f359de09745712a622b6d0c96cadfdcb580f6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:56:37.091202Z","signature_b64":"XHpoEYLbWCRUVDFLZgMqgf3zpwx0xIXHWwwfyO7zqAdlxbLc4EmfbzKDewogOC3zGzJ4vmG8NFnFZRIHq1mgCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9ca988b33e3449600002dddb6a9f359de09745712a622b6d0c96cadfdcb580f6","last_reissued_at":"2026-07-05T11:56:37.090629Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:56:37.090629Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.14704","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:56:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eooCx4Z2LxxotY6A3PvL1lw0I90dT5NdGP3BlzEY+GDD45v7vqFH5sbtJy9nb0Jjr20/uET315qnQED7+7UxAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-28T06:33:27.439828Z"},"content_sha256":"a2b858327ef38c0e45c86fcdb64e8499c97f3fa3e278f70a278d41b9ada5b2ec","schema_version":"1.0","event_id":"sha256:a2b858327ef38c0e45c86fcdb64e8499c97f3fa3e278f70a278d41b9ada5b2ec"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:TSUYRMZ6GREWAAAC3XNWVHZVTX","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MCP-Universe: Benchmarking Large Language Models with Real-World Model Context Protocol Servers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Amrita Saha, Caiming Xiong, Doyen Sahoo, Junnan Li, Prathyusha Jwalapuram, Silvio Savarese, Wenzhuo Yang, Zhiqi Shen, Zirui Zhao, Ziyang Luo","submitted_at":"2025-08-20T13:28:58Z","abstract_excerpt":"The Model Context Protocol has emerged as a transformative standard for connecting large language models to external data sources and tools, rapidly gaining adoption across major AI providers and development platforms. However, existing benchmarks are overly simplistic and fail to capture real application challenges such as long-horizon reasoning and large, unfamiliar tool spaces. To address this critical gap, we introduce MCP-Universe, the first comprehensive benchmark specifically designed to evaluate LLMs in realistic and hard tasks through interaction with real-world MCP servers. Our bench"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.14704","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2508.14704/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:56:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nFsU8dXKDs4SsVb7rqxARgOCPMYsrKAee3jlSdiVib1trI8Ei9AohsJbj5hLpINxSvXxdLCKhKcYTXENmMgRBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-28T06:33:27.440211Z"},"content_sha256":"30407dd0f2ac6be8179abb69df2fb95831284a203c500a1f14f8afcd667e54e1","schema_version":"1.0","event_id":"sha256:30407dd0f2ac6be8179abb69df2fb95831284a203c500a1f14f8afcd667e54e1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TSUYRMZ6GREWAAAC3XNWVHZVTX/bundle.json","state_url":"https://pith.science/pith/TSUYRMZ6GREWAAAC3XNWVHZVTX/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TSUYRMZ6GREWAAAC3XNWVHZVTX/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-28T06:33:27Z","links":{"resolver":"https://pith.science/pith/TSUYRMZ6GREWAAAC3XNWVHZVTX","bundle":"https://pith.science/pith/TSUYRMZ6GREWAAAC3XNWVHZVTX/bundle.json","state":"https://pith.science/pith/TSUYRMZ6GREWAAAC3XNWVHZVTX/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TSUYRMZ6GREWAAAC3XNWVHZVTX/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:TSUYRMZ6GREWAAAC3XNWVHZVTX","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"c5d52610af57d6d96de96d3cf828c17fcdc149f2467f93d986d01688edd0a24e","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-08-20T13:28:58Z","title_canon_sha256":"ee8a8459a50232d35936b97f871d943d5031469396bbf84a27446d89b63f5856"},"schema_version":"1.0","source":{"id":"2508.14704","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.14704","created_at":"2026-07-05T11:56:37Z"},{"alias_kind":"arxiv_version","alias_value":"2508.14704v1","created_at":"2026-07-05T11:56:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.14704","created_at":"2026-07-05T11:56:37Z"},{"alias_kind":"pith_short_12","alias_value":"TSUYRMZ6GREW","created_at":"2026-07-05T11:56:37Z"},{"alias_kind":"pith_short_16","alias_value":"TSUYRMZ6GREWAAAC","created_at":"2026-07-05T11:56:37Z"},{"alias_kind":"pith_short_8","alias_value":"TSUYRMZ6","created_at":"2026-07-05T11:56:37Z"}],"graph_snapshots":[{"event_id":"sha256:30407dd0f2ac6be8179abb69df2fb95831284a203c500a1f14f8afcd667e54e1","target":"graph","created_at":"2026-07-05T11:56:37Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2508.14704/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The Model Context Protocol has emerged as a transformative standard for connecting large language models to external data sources and tools, rapidly gaining adoption across major AI providers and development platforms. However, existing benchmarks are overly simplistic and fail to capture real application challenges such as long-horizon reasoning and large, unfamiliar tool spaces. To address this critical gap, we introduce MCP-Universe, the first comprehensive benchmark specifically designed to evaluate LLMs in realistic and hard tasks through interaction with real-world MCP servers. Our bench","authors_text":"Amrita Saha, Caiming Xiong, Doyen Sahoo, Junnan Li, Prathyusha Jwalapuram, Silvio Savarese, Wenzhuo Yang, Zhiqi Shen, Zirui Zhao, Ziyang Luo","cross_cats":["cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-08-20T13:28:58Z","title":"MCP-Universe: Benchmarking Large Language Models with Real-World Model Context Protocol Servers"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.14704","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:a2b858327ef38c0e45c86fcdb64e8499c97f3fa3e278f70a278d41b9ada5b2ec","target":"record","created_at":"2026-07-05T11:56:37Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"c5d52610af57d6d96de96d3cf828c17fcdc149f2467f93d986d01688edd0a24e","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-08-20T13:28:58Z","title_canon_sha256":"ee8a8459a50232d35936b97f871d943d5031469396bbf84a27446d89b63f5856"},"schema_version":"1.0","source":{"id":"2508.14704","kind":"arxiv","version":1}},"canonical_sha256":"9ca988b33e3449600002dddb6a9f359de09745712a622b6d0c96cadfdcb580f6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9ca988b33e3449600002dddb6a9f359de09745712a622b6d0c96cadfdcb580f6","first_computed_at":"2026-07-05T11:56:37.090629Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:56:37.090629Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"XHpoEYLbWCRUVDFLZgMqgf3zpwx0xIXHWwwfyO7zqAdlxbLc4EmfbzKDewogOC3zGzJ4vmG8NFnFZRIHq1mgCw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:56:37.091202Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.14704","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a2b858327ef38c0e45c86fcdb64e8499c97f3fa3e278f70a278d41b9ada5b2ec","sha256:30407dd0f2ac6be8179abb69df2fb95831284a203c500a1f14f8afcd667e54e1"],"state_sha256":"0afa0db93e1f8a8ab658ad3db31359ee5976556b2244f3ae55abfeea2479a7ed"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"044hIM9E0r5OukVOe6VGjfbAOKqDLth5QvS0K9cLhIgtq07mY4nkb8EjW5HZjMvPH97aO2Tlbs6t0FlZL2/QDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-28T06:33:27.442823Z","bundle_sha256":"4af10053e7c05d7be7b860c1b532bc15b59626d7c325123e2bbe75002b698a1f"}}