{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:FGC62KDMKSXHNWID4YSXYSDNZA","short_pith_number":"pith:FGC62KDM","canonical_record":{"source":{"id":"2507.09063","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2025-07-11T22:45:07Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"84ce3e3b789101fa0298892417416678318b43875e000c1de59d732dbeb67491","abstract_canon_sha256":"b52359ce6828af18f3d3b594aa47eca68dd2bb5c5c00096c84672c1ee2cc4e51"},"schema_version":"1.0"},"canonical_sha256":"2985ed286c54ae76d903e6257c486dc8179d365ddccb4f23070aa9c005934206","source":{"kind":"arxiv","id":"2507.09063","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.09063","created_at":"2026-07-05T11:36:03Z"},{"alias_kind":"arxiv_version","alias_value":"2507.09063v1","created_at":"2026-07-05T11:36:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.09063","created_at":"2026-07-05T11:36:03Z"},{"alias_kind":"pith_short_12","alias_value":"FGC62KDMKSXH","created_at":"2026-07-05T11:36:03Z"},{"alias_kind":"pith_short_16","alias_value":"FGC62KDMKSXHNWID","created_at":"2026-07-05T11:36:03Z"},{"alias_kind":"pith_short_8","alias_value":"FGC62KDM","created_at":"2026-07-05T11:36:03Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:FGC62KDMKSXHNWID4YSXYSDNZA","target":"record","payload":{"canonical_record":{"source":{"id":"2507.09063","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2025-07-11T22:45:07Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"84ce3e3b789101fa0298892417416678318b43875e000c1de59d732dbeb67491","abstract_canon_sha256":"b52359ce6828af18f3d3b594aa47eca68dd2bb5c5c00096c84672c1ee2cc4e51"},"schema_version":"1.0"},"canonical_sha256":"2985ed286c54ae76d903e6257c486dc8179d365ddccb4f23070aa9c005934206","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:36:03.010497Z","signature_b64":"z1EJnzeXSZzfA3kJy1THUy47F211QlMxjMJLeYqmfdzXRRccitUOStVZ9Mzh/v+R0+5V/bY2bSKEbH5em/pMBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2985ed286c54ae76d903e6257c486dc8179d365ddccb4f23070aa9c005934206","last_reissued_at":"2026-07-05T11:36:03.010034Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:36:03.010034Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.09063","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":"2026-07-05T11:36:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bCQcoICeeykRDHM0NHvehH5ff5fsi4pw0eGVPKcATDi0M331oiXRpfeNFMsWSpu8+qDzamgptiMgokiIRf4cCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T04:22:21.118923Z"},"content_sha256":"2a5e029e58d74157c08a855414008645d4b367858b1864a6875c051cf93c7abd","schema_version":"1.0","event_id":"sha256:2a5e029e58d74157c08a855414008645d4b367858b1864a6875c051cf93c7abd"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:FGC62KDMKSXHNWID4YSXYSDNZA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SetupBench: Assessing Software Engineering Agents' Ability to Bootstrap Development Environments","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.SE","authors_text":"Avi Arora, Jinu Jang, Roshanak Zilouchian Moghaddam","submitted_at":"2025-07-11T22:45:07Z","abstract_excerpt":"Modern Large Language Model (LLM) agents promise end to end assistance with real-world software tasks, yet existing benchmarks evaluate LLM agents almost exclusively in pre-baked environments where every dependency is pre-installed. To fill this gap, we introduce SetupBench, a 93 instance benchmark that isolates the environment-bootstrap skill: starting from a bare Linux sandbox, an agent must install packages, resolve dependency conflicts, initialize databases, and configure background services. Our tasks span seven language ecosystems, five database engines, and multi-service orchestration s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.09063","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/2507.09063/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":"2026-07-05T11:36:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"v3t7Y+K2rwdjBeTy/VkRoJlc8kd7KeRRFbl3D0WugoCEqcLg8LjNLrL58Pjr75jbwa5gQ1egwaeJfYv7iKF9Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T04:22:21.119753Z"},"content_sha256":"a2056041180f32c1544463e5aabc1c2dba93dd421304a01f48cdc181ab3534b3","schema_version":"1.0","event_id":"sha256:a2056041180f32c1544463e5aabc1c2dba93dd421304a01f48cdc181ab3534b3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FGC62KDMKSXHNWID4YSXYSDNZA/bundle.json","state_url":"https://pith.science/pith/FGC62KDMKSXHNWID4YSXYSDNZA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FGC62KDMKSXHNWID4YSXYSDNZA/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":"2026-08-07T04:22:21Z","links":{"resolver":"https://pith.science/pith/FGC62KDMKSXHNWID4YSXYSDNZA","bundle":"https://pith.science/pith/FGC62KDMKSXHNWID4YSXYSDNZA/bundle.json","state":"https://pith.science/pith/FGC62KDMKSXHNWID4YSXYSDNZA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FGC62KDMKSXHNWID4YSXYSDNZA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:FGC62KDMKSXHNWID4YSXYSDNZA","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":"b52359ce6828af18f3d3b594aa47eca68dd2bb5c5c00096c84672c1ee2cc4e51","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2025-07-11T22:45:07Z","title_canon_sha256":"84ce3e3b789101fa0298892417416678318b43875e000c1de59d732dbeb67491"},"schema_version":"1.0","source":{"id":"2507.09063","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.09063","created_at":"2026-07-05T11:36:03Z"},{"alias_kind":"arxiv_version","alias_value":"2507.09063v1","created_at":"2026-07-05T11:36:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.09063","created_at":"2026-07-05T11:36:03Z"},{"alias_kind":"pith_short_12","alias_value":"FGC62KDMKSXH","created_at":"2026-07-05T11:36:03Z"},{"alias_kind":"pith_short_16","alias_value":"FGC62KDMKSXHNWID","created_at":"2026-07-05T11:36:03Z"},{"alias_kind":"pith_short_8","alias_value":"FGC62KDM","created_at":"2026-07-05T11:36:03Z"}],"graph_snapshots":[{"event_id":"sha256:a2056041180f32c1544463e5aabc1c2dba93dd421304a01f48cdc181ab3534b3","target":"graph","created_at":"2026-07-05T11:36:03Z","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/2507.09063/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Modern Large Language Model (LLM) agents promise end to end assistance with real-world software tasks, yet existing benchmarks evaluate LLM agents almost exclusively in pre-baked environments where every dependency is pre-installed. To fill this gap, we introduce SetupBench, a 93 instance benchmark that isolates the environment-bootstrap skill: starting from a bare Linux sandbox, an agent must install packages, resolve dependency conflicts, initialize databases, and configure background services. Our tasks span seven language ecosystems, five database engines, and multi-service orchestration s","authors_text":"Avi Arora, Jinu Jang, Roshanak Zilouchian Moghaddam","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2025-07-11T22:45:07Z","title":"SetupBench: Assessing Software Engineering Agents' Ability to Bootstrap Development Environments"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.09063","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":"sha256:2a5e029e58d74157c08a855414008645d4b367858b1864a6875c051cf93c7abd","target":"record","created_at":"2026-07-05T11:36:03Z","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":"b52359ce6828af18f3d3b594aa47eca68dd2bb5c5c00096c84672c1ee2cc4e51","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2025-07-11T22:45:07Z","title_canon_sha256":"84ce3e3b789101fa0298892417416678318b43875e000c1de59d732dbeb67491"},"schema_version":"1.0","source":{"id":"2507.09063","kind":"arxiv","version":1}},"canonical_sha256":"2985ed286c54ae76d903e6257c486dc8179d365ddccb4f23070aa9c005934206","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2985ed286c54ae76d903e6257c486dc8179d365ddccb4f23070aa9c005934206","first_computed_at":"2026-07-05T11:36:03.010034Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:36:03.010034Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"z1EJnzeXSZzfA3kJy1THUy47F211QlMxjMJLeYqmfdzXRRccitUOStVZ9Mzh/v+R0+5V/bY2bSKEbH5em/pMBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:36:03.010497Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.09063","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2a5e029e58d74157c08a855414008645d4b367858b1864a6875c051cf93c7abd","sha256:a2056041180f32c1544463e5aabc1c2dba93dd421304a01f48cdc181ab3534b3"],"state_sha256":"8dd2ba06f6423a43261804c79c7a41de33a790ea5460ffc5d4aa735f95e0cdb1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Wjbx2Aj7023XIzPRbJO9HPd5QTXJgli22jVWLhrjjTcS/g93vVwIedOXKkD2o0a3nSl7VSYoEMmxNDb27WliCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T04:22:21.124385Z","bundle_sha256":"961640159dac9aab7345da6cd874018ec1714cb63738b9214053b37beb9120bd"}}