{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:RHDIGH5IE65MHCSA3A37EDLUQA","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":"363e8ca21e15174c72e05081c081de279e31f39b66662fd97f6560d405c1840b","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-05-13T11:14:59Z","title_canon_sha256":"0912d0135db6157edf6996788b44b8c6fc70f8493c993ae6a8820ef3774e9d02"},"schema_version":"1.0","source":{"id":"2605.13357","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.13357","created_at":"2026-05-18T02:44:48Z"},{"alias_kind":"arxiv_version","alias_value":"2605.13357v1","created_at":"2026-05-18T02:44:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13357","created_at":"2026-05-18T02:44:48Z"},{"alias_kind":"pith_short_12","alias_value":"RHDIGH5IE65M","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"RHDIGH5IE65MHCSA","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"RHDIGH5I","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:73b2800a5a378f7defafb037b8a470fb1e7bdf9ab343ad0751d284e5805ebc7d","target":"graph","created_at":"2026-05-18T02:44:48Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"We propose a different locus: software-engineering capability emerges from a model-harness-environment system, in which a runtime substrate -- the harness -- mediates how a foundation-model agent observes a project, acts on it, receives feedback, and establishes that a change is complete."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That defining eleven component responsibilities and a four-level ladder will systematically improve agent reliability and verifiability, an assumption stated without supporting data or derivation in the abstract."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"The paper defines AI Harness Engineering as a runtime substrate with eleven components and a four-level ladder that reframes agent reliability as a model-harness-environment system property rather than model capability alone."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Software-engineering capability for foundation-model agents emerges from a model-harness-environment system rather than from the model alone."}],"snapshot_sha256":"38745aeb6cb7ad7e0642a2fb3dedd7eacb6f398269ad58f160f8f180eb1fa2d2"},"formal_canon":{"evidence_count":1,"snapshot_sha256":"f4ea6edfd57c9f674ea0a01065b5e807f2e7ea030a40fc70edb6d925cb813dfa"},"paper":{"abstract_excerpt":"Foundation models have transformed automated code generation, yet autonomous software-engineering agents remain unreliable in realistic development settings. The dominant explanation locates this gap in model capability. We propose a different locus: software-engineering capability emerges from a model-harness-environment system, in which a runtime substrate -- the harness -- mediates how a foundation-model agent observes a project, acts on it, receives feedback, and establishes that a change is complete. We formalize this substrate as an AI Harness Engineering and identify eleven component re","authors_text":"Hailin Zhong, Shengxin Zhu","cross_cats":["cs.AI"],"headline":"Software-engineering capability for foundation-model agents emerges from a model-harness-environment system rather than from the model alone.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-05-13T11:14:59Z","title":"AI Harness Engineering: A Runtime Substrate for Foundation-Model Software Agents"},"references":{"count":18,"internal_anchors":2,"resolved_work":18,"sample":[{"cited_arxiv_id":"2107.03374","doi":"","is_internal_anchor":true,"ref_index":1,"title":"Evaluating Large Language Models Trained on Code","work_id":"042493e9-b26f-4b4e-bbde-382072ca9b08","year":2021},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared Kaplan, Prafulla Dhariwal, et al","work_id":"857e276a-33ad-44d4-9d43-c728bd2db249","year":2020},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Jimenez, John Yang, Alexander Wettig, Shunyu Yao, Kexin Pei, Ofir Press, and Karthik Narasimhan","work_id":"faab4af1-0cb8-4520-9a0a-c7a1a55118e4","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Jimenez, Alexander Wettig, Kilian Lieret, Shunyu Yao, Karthik Narasimhan, and Ofir Press","work_id":"7a8ddf3e-1730-4670-9cf2-8789e3706977","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Xu, Xiangru Tang, Mingchen Zhuge, Jiayi Pan, Yueqi Song, Bowen Li, Jaskirat Singh, Hoang H","work_id":"ba783dfe-165d-4d63-b7f9-9708af7040b3","year":2025}],"snapshot_sha256":"92dab86cc85b5856542c23b8222e918e76631680ad80ee1ff237fcbb35bbd41b"},"source":{"id":"2605.13357","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-14T18:10:33.629110Z","id":"f734b0c7-8b0d-4bdd-9d7c-aac713be3f00","model_set":{"reader":"grok-4.3"},"one_line_summary":"The paper defines AI Harness Engineering as a runtime substrate with eleven components and a four-level ladder that reframes agent reliability as a model-harness-environment system property rather than model capability alone.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Software-engineering capability for foundation-model agents emerges from a model-harness-environment system rather than from the model alone.","strongest_claim":"We propose a different locus: software-engineering capability emerges from a model-harness-environment system, in which a runtime substrate -- the harness -- mediates how a foundation-model agent observes a project, acts on it, receives feedback, and establishes that a change is complete.","weakest_assumption":"That defining eleven component responsibilities and a four-level ladder will systematically improve agent reliability and verifiability, an assumption stated without supporting data or derivation in the abstract."}},"verdict_id":"f734b0c7-8b0d-4bdd-9d7c-aac713be3f00"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:5efd0a8aa8761e3e5421ffba81241544d36037787bafc7aa86240e77ee1b1f3c","target":"record","created_at":"2026-05-18T02:44:48Z","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":"363e8ca21e15174c72e05081c081de279e31f39b66662fd97f6560d405c1840b","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-05-13T11:14:59Z","title_canon_sha256":"0912d0135db6157edf6996788b44b8c6fc70f8493c993ae6a8820ef3774e9d02"},"schema_version":"1.0","source":{"id":"2605.13357","kind":"arxiv","version":1}},"canonical_sha256":"89c6831fa827bac38a40d837f20d74801a0673750c947c322a7e0d895a918d29","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"89c6831fa827bac38a40d837f20d74801a0673750c947c322a7e0d895a918d29","first_computed_at":"2026-05-18T02:44:48.199273Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T02:44:48.199273Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"e9qdhJ+qQJ6g1R4S3iaWork2YCgzm25isnwmM5oQmQEODbGEpS/czFYKH4Sy2TKLHm7uzl0T+MM+IUdZTB19BA==","signature_status":"signed_v1","signed_at":"2026-05-18T02:44:48.199695Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.13357","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5efd0a8aa8761e3e5421ffba81241544d36037787bafc7aa86240e77ee1b1f3c","sha256:73b2800a5a378f7defafb037b8a470fb1e7bdf9ab343ad0751d284e5805ebc7d"],"state_sha256":"ea73a288b7c760b54374d5426365f14b9e36411415715d1ce50c66fba32d0175"}