{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:NU7HM3TM5CSVZWTMUOBHPBB6TC","short_pith_number":"pith:NU7HM3TM","schema_version":"1.0","canonical_sha256":"6d3e766e6ce8a55cda6ca38277843e98b968c0adc1c4d73cfc62f582b8ff1811","source":{"kind":"arxiv","id":"2607.22585","version":1},"attestation_state":"computed","paper":{"title":"The Scaffold Effect in Coding Agents: Harness Choice as a Hidden Variable in Coding-Agent Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Naman Vats, Oleg Golev","submitted_at":"2026-06-08T20:27:06Z","abstract_excerpt":"Public leaderboards for coding agents typically rank systems by model name and pass rate, while the surrounding harness (the scaffold that issues tools, manages context, and decides when to stop) is often under-specified. Model-to-model comparison is valid when the harness is fixed; when it varies, performance and efficiency conflate model and scaffold effects. We evaluate Qwen 3.6 Plus and MiniMax M2.5 across three open-source harnesses (Goose, OpenCode, OpenHands-SDK) on a stratified 50-task subset of Terminal-Bench Pro. Harness choice induces up to a 40x difference in tokens per solved task"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2607.22585","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-08T20:27:06Z","cross_cats_sorted":[],"title_canon_sha256":"eddb65842dcf31dfc5c64bd059f6ebf20141aadbe421b6a73c6540dfe98705e1","abstract_canon_sha256":"c1734da34455b4976fcb4edac0224ca0a604d478412711ae2d062ade14a4eac9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-28T00:21:43.547747Z","signature_b64":"3hCKtRTGJsUDb0XHuCb5YJbpEudbTeEwDHJ5wc/ao+BAZLxR2bdlWjkX5KaRDIg0NVb/DJjGx1Ml/pY2fAUOAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6d3e766e6ce8a55cda6ca38277843e98b968c0adc1c4d73cfc62f582b8ff1811","last_reissued_at":"2026-07-28T00:21:43.546823Z","signature_status":"signed_v1","first_computed_at":"2026-07-28T00:21:43.546823Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Scaffold Effect in Coding Agents: Harness Choice as a Hidden Variable in Coding-Agent Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Naman Vats, Oleg Golev","submitted_at":"2026-06-08T20:27:06Z","abstract_excerpt":"Public leaderboards for coding agents typically rank systems by model name and pass rate, while the surrounding harness (the scaffold that issues tools, manages context, and decides when to stop) is often under-specified. Model-to-model comparison is valid when the harness is fixed; when it varies, performance and efficiency conflate model and scaffold effects. We evaluate Qwen 3.6 Plus and MiniMax M2.5 across three open-source harnesses (Goose, OpenCode, OpenHands-SDK) on a stratified 50-task subset of Terminal-Bench Pro. Harness choice induces up to a 40x difference in tokens per solved task"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.22585","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/2607.22585/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2607.22585","created_at":"2026-07-28T00:21:43.547248+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.22585v1","created_at":"2026-07-28T00:21:43.547248+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.22585","created_at":"2026-07-28T00:21:43.547248+00:00"},{"alias_kind":"pith_short_12","alias_value":"NU7HM3TM5CSV","created_at":"2026-07-28T00:21:43.547248+00:00"},{"alias_kind":"pith_short_16","alias_value":"NU7HM3TM5CSVZWTM","created_at":"2026-07-28T00:21:43.547248+00:00"},{"alias_kind":"pith_short_8","alias_value":"NU7HM3TM","created_at":"2026-07-28T00:21:43.547248+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2608.09480","citing_title":"From Prompt to Harness: Coderlet from Scratch","ref_index":8,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NU7HM3TM5CSVZWTMUOBHPBB6TC","json":"https://pith.science/pith/NU7HM3TM5CSVZWTMUOBHPBB6TC.json","graph_json":"https://pith.science/api/pith-number/NU7HM3TM5CSVZWTMUOBHPBB6TC/graph.json","events_json":"https://pith.science/api/pith-number/NU7HM3TM5CSVZWTMUOBHPBB6TC/events.json","paper":"https://pith.science/paper/NU7HM3TM"},"agent_actions":{"view_html":"https://pith.science/pith/NU7HM3TM5CSVZWTMUOBHPBB6TC","download_json":"https://pith.science/pith/NU7HM3TM5CSVZWTMUOBHPBB6TC.json","view_paper":"https://pith.science/paper/NU7HM3TM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.22585&json=true","fetch_graph":"https://pith.science/api/pith-number/NU7HM3TM5CSVZWTMUOBHPBB6TC/graph.json","fetch_events":"https://pith.science/api/pith-number/NU7HM3TM5CSVZWTMUOBHPBB6TC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NU7HM3TM5CSVZWTMUOBHPBB6TC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NU7HM3TM5CSVZWTMUOBHPBB6TC/action/storage_attestation","attest_author":"https://pith.science/pith/NU7HM3TM5CSVZWTMUOBHPBB6TC/action/author_attestation","sign_citation":"https://pith.science/pith/NU7HM3TM5CSVZWTMUOBHPBB6TC/action/citation_signature","submit_replication":"https://pith.science/pith/NU7HM3TM5CSVZWTMUOBHPBB6TC/action/replication_record"}},"created_at":"2026-07-28T00:21:43.547248+00:00","updated_at":"2026-07-28T00:21:43.547248+00:00"}