{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6ONVFNPI3ZGS76LMCWLCTYR5MJ","short_pith_number":"pith:6ONVFNPI","schema_version":"1.0","canonical_sha256":"f39b52b5e8de4d2ff96c159629e23d625e1dc1ec86997f97438faaf942ee54ef","source":{"kind":"arxiv","id":"2604.25850","version":4},"attestation_state":"computed","paper":{"title":"Agentic Harness Engineering: Observability-Driven Automatic Evolution of Coding-Agent Harnesses","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Three observability pillars let coding-agent harnesses evolve autonomously to beat human designs and transfer across benchmarks.","cross_cats":["cs.SE"],"primary_cat":"cs.CL","authors_text":"Chengjun Pan, Hang Yan, Jiahang Lin, Lizhi Lin, Shichun Liu, Shihan Dou, Tao Gui, Xuanjing Huang, Yu-Gang Jiang, Zhenhua Han, Zhiheng Xi","submitted_at":"2026-04-28T16:55:02Z","abstract_excerpt":"Harnesses are now central to coding-agent performance, mediating how models interact with tools and execution environments. Yet harness engineering remains a manual craft, because automating it faces a heterogeneous action space across editable components, voluminous trajectories that bury actionable signal, and edits whose effect is hard to attribute. We introduce Agentic Harness Engineering (AHE), a closed loop that addresses these challenges through three matched observability pillars: (1) component observability gives every editable harness component a file-level representation so the acti"},"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":"2604.25850","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-04-28T16:55:02Z","cross_cats_sorted":["cs.SE"],"title_canon_sha256":"c7b83e7ad798a3955b2ec3441ac7ba8f62c56efb704cc3f89b7bf7fc8e01046f","abstract_canon_sha256":"5bbdee89bffc3cd0212d2d5031c2ed5765e647efb58f9c1cd03725e88a9fcd4e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:05:45.402259Z","signature_b64":"ayKM+MK2fE7ADMXHqBY1pifaIU/4G1nPxNWuQhASB3ReG3PtJqdiKG8nbVO3ejnRC0GmzCxPKW2R3wbRxFaBBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f39b52b5e8de4d2ff96c159629e23d625e1dc1ec86997f97438faaf942ee54ef","last_reissued_at":"2026-05-20T00:05:45.401710Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:05:45.401710Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Agentic Harness Engineering: Observability-Driven Automatic Evolution of Coding-Agent Harnesses","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Three observability pillars let coding-agent harnesses evolve autonomously to beat human designs and transfer across benchmarks.","cross_cats":["cs.SE"],"primary_cat":"cs.CL","authors_text":"Chengjun Pan, Hang Yan, Jiahang Lin, Lizhi Lin, Shichun Liu, Shihan Dou, Tao Gui, Xuanjing Huang, Yu-Gang Jiang, Zhenhua Han, Zhiheng Xi","submitted_at":"2026-04-28T16:55:02Z","abstract_excerpt":"Harnesses are now central to coding-agent performance, mediating how models interact with tools and execution environments. Yet harness engineering remains a manual craft, because automating it faces a heterogeneous action space across editable components, voluminous trajectories that bury actionable signal, and edits whose effect is hard to attribute. We introduce Agentic Harness Engineering (AHE), a closed loop that addresses these challenges through three matched observability pillars: (1) component observability gives every editable harness component a file-level representation so the acti"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"ten AHE iterations lift pass@1 on Terminal-Bench 2 from 69.7% to 77.0%, surpassing the human-designed harness Codex-CLI (71.9%) and the self-evolving baselines ACE and TF-GRPO. The frozen harness transfers without re-evolution: on SWE-bench-verified it tops aggregate success at 12% fewer tokens than the seed, and on Terminal-Bench 2 it yields +5.1 to +10.1pp cross-family gains across three alternate model families.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the three observability pillars sufficiently constrain the action space and provide actionable signal so that the evolution loop produces generalizable improvements rather than benchmark-specific overfitting or noise-driven changes.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"AHE automates coding-agent harness evolution via component, experience, and decision observability, raising Terminal-Bench 2 pass@1 from 69.7% to 77.0% with transfer gains across models and benchmarks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Three observability pillars let coding-agent harnesses evolve autonomously to beat human designs and transfer across benchmarks.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"a2e5b0e9c835855ba0d59c28e90c9105e6239d1a8af3b112cca51ebd8cb8a7c5"},"source":{"id":"2604.25850","kind":"arxiv","version":4},"verdict":{"id":"887645bd-c5cc-42e6-b8be-6c0dde0800bc","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-07T16:12:58.518385Z","strongest_claim":"ten AHE iterations lift pass@1 on Terminal-Bench 2 from 69.7% to 77.0%, surpassing the human-designed harness Codex-CLI (71.9%) and the self-evolving baselines ACE and TF-GRPO. The frozen harness transfers without re-evolution: on SWE-bench-verified it tops aggregate success at 12% fewer tokens than the seed, and on Terminal-Bench 2 it yields +5.1 to +10.1pp cross-family gains across three alternate model families.","one_line_summary":"AHE automates coding-agent harness evolution via component, experience, and decision observability, raising Terminal-Bench 2 pass@1 from 69.7% to 77.0% with transfer gains across models and benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the three observability pillars sufficiently constrain the action space and provide actionable signal so that the evolution loop produces generalizable improvements rather than benchmark-specific overfitting or noise-driven changes.","pith_extraction_headline":"Three observability pillars let coding-agent harnesses evolve autonomously to beat human designs and transfer across benchmarks."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.25850/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_compliance","ran_at":"2026-05-19T20:43:22.071216Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"ea2b183ffce07ff7ac02fc3848eb521e69fa16beceae61e1f461ea05cac388f6"},"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":"2604.25850","created_at":"2026-05-20T00:05:45.401809+00:00"},{"alias_kind":"arxiv_version","alias_value":"2604.25850v4","created_at":"2026-05-20T00:05:45.401809+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.25850","created_at":"2026-05-20T00:05:45.401809+00:00"},{"alias_kind":"pith_short_12","alias_value":"6ONVFNPI3ZGS","created_at":"2026-05-20T00:05:45.401809+00:00"},{"alias_kind":"pith_short_16","alias_value":"6ONVFNPI3ZGS76LM","created_at":"2026-05-20T00:05:45.401809+00:00"},{"alias_kind":"pith_short_8","alias_value":"6ONVFNPI","created_at":"2026-05-20T00:05:45.401809+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"2605.18401","citing_title":"SkillsVote: Lifecycle Governance of Agent Skills from Collection, Recommendation to Evolution","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18747","citing_title":"Code as Agent Harness","ref_index":286,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18476","citing_title":"AI4BayesCode: From Natural Language Descriptions to Validated Modular Stateful Bayesian Samplers","ref_index":74,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6ONVFNPI3ZGS76LMCWLCTYR5MJ","json":"https://pith.science/pith/6ONVFNPI3ZGS76LMCWLCTYR5MJ.json","graph_json":"https://pith.science/api/pith-number/6ONVFNPI3ZGS76LMCWLCTYR5MJ/graph.json","events_json":"https://pith.science/api/pith-number/6ONVFNPI3ZGS76LMCWLCTYR5MJ/events.json","paper":"https://pith.science/paper/6ONVFNPI"},"agent_actions":{"view_html":"https://pith.science/pith/6ONVFNPI3ZGS76LMCWLCTYR5MJ","download_json":"https://pith.science/pith/6ONVFNPI3ZGS76LMCWLCTYR5MJ.json","view_paper":"https://pith.science/paper/6ONVFNPI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2604.25850&json=true","fetch_graph":"https://pith.science/api/pith-number/6ONVFNPI3ZGS76LMCWLCTYR5MJ/graph.json","fetch_events":"https://pith.science/api/pith-number/6ONVFNPI3ZGS76LMCWLCTYR5MJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6ONVFNPI3ZGS76LMCWLCTYR5MJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6ONVFNPI3ZGS76LMCWLCTYR5MJ/action/storage_attestation","attest_author":"https://pith.science/pith/6ONVFNPI3ZGS76LMCWLCTYR5MJ/action/author_attestation","sign_citation":"https://pith.science/pith/6ONVFNPI3ZGS76LMCWLCTYR5MJ/action/citation_signature","submit_replication":"https://pith.science/pith/6ONVFNPI3ZGS76LMCWLCTYR5MJ/action/replication_record"}},"created_at":"2026-05-20T00:05:45.401809+00:00","updated_at":"2026-05-20T00:05:45.401809+00:00"}