{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:HFETZMBQYJ4XW3ASQPFUAJE77A","short_pith_number":"pith:HFETZMBQ","canonical_record":{"source":{"id":"2608.05263","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-08-05T17:27:15Z","cross_cats_sorted":[],"title_canon_sha256":"6adaddc57b3f66a113891a96caaa837e2870d044f915ba4f005bed9724a6be25","abstract_canon_sha256":"3eb39a519db37561290da42e11fd3ebe0eaecc53fefdd912a0e14752e21f18f2"},"schema_version":"1.0"},"canonical_sha256":"39493cb030c2797b6c1283cb40249ff80c428d5c5cf4db3e0df6afb5f8c4cfb2","source":{"kind":"arxiv","id":"2608.05263","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.05263","created_at":"2026-08-07T00:46:38Z"},{"alias_kind":"arxiv_version","alias_value":"2608.05263v1","created_at":"2026-08-07T00:46:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.05263","created_at":"2026-08-07T00:46:38Z"},{"alias_kind":"pith_short_12","alias_value":"HFETZMBQYJ4X","created_at":"2026-08-07T00:46:38Z"},{"alias_kind":"pith_short_16","alias_value":"HFETZMBQYJ4XW3AS","created_at":"2026-08-07T00:46:38Z"},{"alias_kind":"pith_short_8","alias_value":"HFETZMBQ","created_at":"2026-08-07T00:46:38Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:HFETZMBQYJ4XW3ASQPFUAJE77A","target":"record","payload":{"canonical_record":{"source":{"id":"2608.05263","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-08-05T17:27:15Z","cross_cats_sorted":[],"title_canon_sha256":"6adaddc57b3f66a113891a96caaa837e2870d044f915ba4f005bed9724a6be25","abstract_canon_sha256":"3eb39a519db37561290da42e11fd3ebe0eaecc53fefdd912a0e14752e21f18f2"},"schema_version":"1.0"},"canonical_sha256":"39493cb030c2797b6c1283cb40249ff80c428d5c5cf4db3e0df6afb5f8c4cfb2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-07T00:46:38.172062Z","signature_b64":"77+110+tBcKGCC0Pi55TPzw+3NDgPSkERHwNeCBDNFsDGqoaX6WXoikDWBK9IyiV53I/Crfaxs4R4DRvTh1EAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"39493cb030c2797b6c1283cb40249ff80c428d5c5cf4db3e0df6afb5f8c4cfb2","last_reissued_at":"2026-08-07T00:46:38.170678Z","signature_status":"signed_v1","first_computed_at":"2026-08-07T00:46:38.170678Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2608.05263","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":"2026-08-07T00:46:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/+I5kQw+4S3aPhle91pEZXz50se6GpkQOUzZb2YtgK5moBCoU2OwUUoWAX7XYvLQ2bRDUpOMMOGIPtWXk0DtDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T00:34:47.694484Z"},"content_sha256":"941f978d458c1a6ccbce3da611b2b430b4a6143dbdd9d807f1f01f78e1e0cfe7","schema_version":"1.0","event_id":"sha256:941f978d458c1a6ccbce3da611b2b430b4a6143dbdd9d807f1f01f78e1e0cfe7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:HFETZMBQYJ4XW3ASQPFUAJE77A","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"OrchestraBench: Evaluating Multi-Agent Orchestration Failure Modes, Recovery, and Decomposition Quality","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Natan Vidra, Sharon Zheng, Spurthi Setty, Yidian Chen, Yingzi Gu","submitted_at":"2026-08-05T17:27:15Z","abstract_excerpt":"Multi-agent orchestration frameworks are moving from demos to production, yet benchmarks typically report task accuracy without diagnosing why a pipeline failed, where a cascade began, or which routing decision caused the breakdown. OrchestraBench evaluates failure, recovery, and decomposition through a controlled, seed-reproducible failure-injection harness over templated enterprise workflows. It introduces cascade radius and per-failure-mode recovery as primary metrics and compares routing policies with bootstrap confidence intervals and paired tests. On a 26-case gold-labelled diagnostic, a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.05263","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/2608.05263/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":"2026-08-07T00:46:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"krR+R0Nd4A3K5o4mSh6lYKnKcIV52F14eHcQRHu5k3xkhsNh0N4iosDw9RIhpNCpjUx73L0yQtI+cXOhMlFZAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T00:34:47.695731Z"},"content_sha256":"bb5576c9e174a345f26c64322a0209ad7bd382527db388a09514c8d77ae11f39","schema_version":"1.0","event_id":"sha256:bb5576c9e174a345f26c64322a0209ad7bd382527db388a09514c8d77ae11f39"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HFETZMBQYJ4XW3ASQPFUAJE77A/bundle.json","state_url":"https://pith.science/pith/HFETZMBQYJ4XW3ASQPFUAJE77A/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HFETZMBQYJ4XW3ASQPFUAJE77A/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":"2026-08-14T00:34:47Z","links":{"resolver":"https://pith.science/pith/HFETZMBQYJ4XW3ASQPFUAJE77A","bundle":"https://pith.science/pith/HFETZMBQYJ4XW3ASQPFUAJE77A/bundle.json","state":"https://pith.science/pith/HFETZMBQYJ4XW3ASQPFUAJE77A/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HFETZMBQYJ4XW3ASQPFUAJE77A/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:HFETZMBQYJ4XW3ASQPFUAJE77A","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":"3eb39a519db37561290da42e11fd3ebe0eaecc53fefdd912a0e14752e21f18f2","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-08-05T17:27:15Z","title_canon_sha256":"6adaddc57b3f66a113891a96caaa837e2870d044f915ba4f005bed9724a6be25"},"schema_version":"1.0","source":{"id":"2608.05263","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.05263","created_at":"2026-08-07T00:46:38Z"},{"alias_kind":"arxiv_version","alias_value":"2608.05263v1","created_at":"2026-08-07T00:46:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.05263","created_at":"2026-08-07T00:46:38Z"},{"alias_kind":"pith_short_12","alias_value":"HFETZMBQYJ4X","created_at":"2026-08-07T00:46:38Z"},{"alias_kind":"pith_short_16","alias_value":"HFETZMBQYJ4XW3AS","created_at":"2026-08-07T00:46:38Z"},{"alias_kind":"pith_short_8","alias_value":"HFETZMBQ","created_at":"2026-08-07T00:46:38Z"}],"graph_snapshots":[{"event_id":"sha256:bb5576c9e174a345f26c64322a0209ad7bd382527db388a09514c8d77ae11f39","target":"graph","created_at":"2026-08-07T00:46:38Z","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/2608.05263/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multi-agent orchestration frameworks are moving from demos to production, yet benchmarks typically report task accuracy without diagnosing why a pipeline failed, where a cascade began, or which routing decision caused the breakdown. OrchestraBench evaluates failure, recovery, and decomposition through a controlled, seed-reproducible failure-injection harness over templated enterprise workflows. It introduces cascade radius and per-failure-mode recovery as primary metrics and compares routing policies with bootstrap confidence intervals and paired tests. On a 26-case gold-labelled diagnostic, a","authors_text":"Natan Vidra, Sharon Zheng, Spurthi Setty, Yidian Chen, Yingzi Gu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-08-05T17:27:15Z","title":"OrchestraBench: Evaluating Multi-Agent Orchestration Failure Modes, Recovery, and Decomposition Quality"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.05263","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":"sha256:941f978d458c1a6ccbce3da611b2b430b4a6143dbdd9d807f1f01f78e1e0cfe7","target":"record","created_at":"2026-08-07T00:46:38Z","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":"3eb39a519db37561290da42e11fd3ebe0eaecc53fefdd912a0e14752e21f18f2","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-08-05T17:27:15Z","title_canon_sha256":"6adaddc57b3f66a113891a96caaa837e2870d044f915ba4f005bed9724a6be25"},"schema_version":"1.0","source":{"id":"2608.05263","kind":"arxiv","version":1}},"canonical_sha256":"39493cb030c2797b6c1283cb40249ff80c428d5c5cf4db3e0df6afb5f8c4cfb2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"39493cb030c2797b6c1283cb40249ff80c428d5c5cf4db3e0df6afb5f8c4cfb2","first_computed_at":"2026-08-07T00:46:38.170678Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-07T00:46:38.170678Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"77+110+tBcKGCC0Pi55TPzw+3NDgPSkERHwNeCBDNFsDGqoaX6WXoikDWBK9IyiV53I/Crfaxs4R4DRvTh1EAA==","signature_status":"signed_v1","signed_at":"2026-08-07T00:46:38.172062Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.05263","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:941f978d458c1a6ccbce3da611b2b430b4a6143dbdd9d807f1f01f78e1e0cfe7","sha256:bb5576c9e174a345f26c64322a0209ad7bd382527db388a09514c8d77ae11f39"],"state_sha256":"83a026afce4c64f23d94aaa62202075df21bf563d84bd6291ebadc6861a47961"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CwOVxFY8DsO0IXbu8nyu5bmNSRaEqSnfqGRcaYCNzqDmHbjG3+uMng125lql91VV63tkt9iw0BTyg+zfN+z5Cw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T00:34:47.792693Z","bundle_sha256":"48a64a1f783c92b28238383af5219c7b61d7366c4a839d23a9cbdc99112b852f"}}