{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:GON6CRZ5DKZPOPTD6K2UXDPXRB","short_pith_number":"pith:GON6CRZ5","canonical_record":{"source":{"id":"2607.14642","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-16T07:09:49Z","cross_cats_sorted":["cs.SE"],"title_canon_sha256":"9c35675f6b6622f3515429375062efed42753d9df3ac309e00a2a3db4044b098","abstract_canon_sha256":"6509fbebbc6d0e5ffb5c840ab20350d89d57483ebd6c887f5a5fff46002aa8bb"},"schema_version":"1.0"},"canonical_sha256":"339be1473d1ab2f73e63f2b54b8df78866a1637bf9ad38fb719e42555ca62b9c","source":{"kind":"arxiv","id":"2607.14642","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.14642","created_at":"2026-07-17T01:21:22Z"},{"alias_kind":"arxiv_version","alias_value":"2607.14642v1","created_at":"2026-07-17T01:21:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.14642","created_at":"2026-07-17T01:21:22Z"},{"alias_kind":"pith_short_12","alias_value":"GON6CRZ5DKZP","created_at":"2026-07-17T01:21:22Z"},{"alias_kind":"pith_short_16","alias_value":"GON6CRZ5DKZPOPTD","created_at":"2026-07-17T01:21:22Z"},{"alias_kind":"pith_short_8","alias_value":"GON6CRZ5","created_at":"2026-07-17T01:21:22Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:GON6CRZ5DKZPOPTD6K2UXDPXRB","target":"record","payload":{"canonical_record":{"source":{"id":"2607.14642","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-16T07:09:49Z","cross_cats_sorted":["cs.SE"],"title_canon_sha256":"9c35675f6b6622f3515429375062efed42753d9df3ac309e00a2a3db4044b098","abstract_canon_sha256":"6509fbebbc6d0e5ffb5c840ab20350d89d57483ebd6c887f5a5fff46002aa8bb"},"schema_version":"1.0"},"canonical_sha256":"339be1473d1ab2f73e63f2b54b8df78866a1637bf9ad38fb719e42555ca62b9c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-17T01:21:22.486803Z","signature_b64":"ESKSXowCH9lItMucBgfJY2YkrQhLYIvGMcTep5yG7TyZjq9xIxBJkrLsUPSEbRWRrhiN8J0KGsx52cH7fXXaCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"339be1473d1ab2f73e63f2b54b8df78866a1637bf9ad38fb719e42555ca62b9c","last_reissued_at":"2026-07-17T01:21:22.485905Z","signature_status":"signed_v1","first_computed_at":"2026-07-17T01:21:22.485905Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.14642","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":"2026-07-17T01:21:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"A6iPAsl1nEY+xxPtweAkEZepx0DGyu+93mpwx/VdVRlCKS2dhqWnvs9RW5UB+asn1Fbacl0gSFwE9bmSx3QQAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-31T14:38:38.117872Z"},"content_sha256":"b3b65a084c2143c51790d5aaab613fb3dd49578456b51479b1384f476c05bb44","schema_version":"1.0","event_id":"sha256:b3b65a084c2143c51790d5aaab613fb3dd49578456b51479b1384f476c05bb44"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:GON6CRZ5DKZPOPTD6K2UXDPXRB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MCPEvol-Bench: Benchmarking LLM Agent Performance Across Dynamic Evolutions of MCP Servers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SE"],"primary_cat":"cs.AI","authors_text":"Bo Ding, Dawei Feng, Huaimin Wang, Huanxi Liu, Jiaqi Liao, Kun Hu, Pengfei Qian, Qiang Wang, Yuanzhao Zhai","submitted_at":"2026-07-16T07:09:49Z","abstract_excerpt":"As Model Context Protocol (MCP) servers emerge as the core infrastructure for connecting LLMs with external tools, existing benchmarks leverage real-world MCP servers to evaluate LLM agents' tool-using capabilities. However, these benchmarks overlook the continuous evolution of tool interfaces and functionalities within MCP servers, resulting in flawed assessments that fail to capture the agent's adaptability in changing tool landscapes. To bridge this gap, we introduce \\textbf{MCPEvol-Bench}, a novel benchmark for evaluating the task-solving capabilities of LLM agents under dynamic toolset ev"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.14642","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/2607.14642/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":"2026-07-17T01:21:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"h425PjwPdeKx2VGNCU2fTiGoP+8V60Jauk0QvZVMNoVTYYqAFn2e29I9R7NcrBBQaLSZXVFV0KQCCVzJ1HZ9Dw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-31T14:38:38.118195Z"},"content_sha256":"59cd478e4c760b9f5a84ea37229c55578187dcee775581e0d847d688fb5603dc","schema_version":"1.0","event_id":"sha256:59cd478e4c760b9f5a84ea37229c55578187dcee775581e0d847d688fb5603dc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GON6CRZ5DKZPOPTD6K2UXDPXRB/bundle.json","state_url":"https://pith.science/pith/GON6CRZ5DKZPOPTD6K2UXDPXRB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GON6CRZ5DKZPOPTD6K2UXDPXRB/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":"2026-07-31T14:38:38Z","links":{"resolver":"https://pith.science/pith/GON6CRZ5DKZPOPTD6K2UXDPXRB","bundle":"https://pith.science/pith/GON6CRZ5DKZPOPTD6K2UXDPXRB/bundle.json","state":"https://pith.science/pith/GON6CRZ5DKZPOPTD6K2UXDPXRB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GON6CRZ5DKZPOPTD6K2UXDPXRB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:GON6CRZ5DKZPOPTD6K2UXDPXRB","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":"6509fbebbc6d0e5ffb5c840ab20350d89d57483ebd6c887f5a5fff46002aa8bb","cross_cats_sorted":["cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-16T07:09:49Z","title_canon_sha256":"9c35675f6b6622f3515429375062efed42753d9df3ac309e00a2a3db4044b098"},"schema_version":"1.0","source":{"id":"2607.14642","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.14642","created_at":"2026-07-17T01:21:22Z"},{"alias_kind":"arxiv_version","alias_value":"2607.14642v1","created_at":"2026-07-17T01:21:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.14642","created_at":"2026-07-17T01:21:22Z"},{"alias_kind":"pith_short_12","alias_value":"GON6CRZ5DKZP","created_at":"2026-07-17T01:21:22Z"},{"alias_kind":"pith_short_16","alias_value":"GON6CRZ5DKZPOPTD","created_at":"2026-07-17T01:21:22Z"},{"alias_kind":"pith_short_8","alias_value":"GON6CRZ5","created_at":"2026-07-17T01:21:22Z"}],"graph_snapshots":[{"event_id":"sha256:59cd478e4c760b9f5a84ea37229c55578187dcee775581e0d847d688fb5603dc","target":"graph","created_at":"2026-07-17T01:21:22Z","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/2607.14642/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As Model Context Protocol (MCP) servers emerge as the core infrastructure for connecting LLMs with external tools, existing benchmarks leverage real-world MCP servers to evaluate LLM agents' tool-using capabilities. However, these benchmarks overlook the continuous evolution of tool interfaces and functionalities within MCP servers, resulting in flawed assessments that fail to capture the agent's adaptability in changing tool landscapes. To bridge this gap, we introduce \\textbf{MCPEvol-Bench}, a novel benchmark for evaluating the task-solving capabilities of LLM agents under dynamic toolset ev","authors_text":"Bo Ding, Dawei Feng, Huaimin Wang, Huanxi Liu, Jiaqi Liao, Kun Hu, Pengfei Qian, Qiang Wang, Yuanzhao Zhai","cross_cats":["cs.SE"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-16T07:09:49Z","title":"MCPEvol-Bench: Benchmarking LLM Agent Performance Across Dynamic Evolutions of MCP Servers"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.14642","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:b3b65a084c2143c51790d5aaab613fb3dd49578456b51479b1384f476c05bb44","target":"record","created_at":"2026-07-17T01:21:22Z","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":"6509fbebbc6d0e5ffb5c840ab20350d89d57483ebd6c887f5a5fff46002aa8bb","cross_cats_sorted":["cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-16T07:09:49Z","title_canon_sha256":"9c35675f6b6622f3515429375062efed42753d9df3ac309e00a2a3db4044b098"},"schema_version":"1.0","source":{"id":"2607.14642","kind":"arxiv","version":1}},"canonical_sha256":"339be1473d1ab2f73e63f2b54b8df78866a1637bf9ad38fb719e42555ca62b9c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"339be1473d1ab2f73e63f2b54b8df78866a1637bf9ad38fb719e42555ca62b9c","first_computed_at":"2026-07-17T01:21:22.485905Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-17T01:21:22.485905Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ESKSXowCH9lItMucBgfJY2YkrQhLYIvGMcTep5yG7TyZjq9xIxBJkrLsUPSEbRWRrhiN8J0KGsx52cH7fXXaCw==","signature_status":"signed_v1","signed_at":"2026-07-17T01:21:22.486803Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.14642","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b3b65a084c2143c51790d5aaab613fb3dd49578456b51479b1384f476c05bb44","sha256:59cd478e4c760b9f5a84ea37229c55578187dcee775581e0d847d688fb5603dc"],"state_sha256":"0afb41457d3c1ce7a329eaf57c42945281ca6cb8a35ca8621bf7c986d731a928"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pSO2llD8ma6Kvz9UFufO4jYLZpUnQBOFegcm6jDczggKaA5sOHwU4GMiof0flO8hlpPdqwWbGQvwS8KQxivDBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-31T14:38:38.120541Z","bundle_sha256":"1f53f0404f66927628b1af243298bd1ed3229aa5f34bd891a44d5fd856d81bbd"}}