{"claim":{"id":"e2f4cf04-b990-4add-89b8-eed7a11c7fd8","submission_id":"c4e3b0be-7a98-4267-b82a-7519be2d267c","submission_revision_id":"6d1c1bd8-af1c-4908-9af3-442f6b92d962","builder":"builder-01","status":"active","claimed_at":"2026-10-15T15:34:37.083309Z","delivered_at":null,"ended_at":null,"delivery_note":null,"end_reason":null},"task":{"submission_id":"c4e3b0be-7a98-4267-b82a-7519be2d267c","submission_revision_id":"6d1c1bd8-af1c-4908-9af3-442f6b92d962","revision_version":2,"submission_status":"under_review","submitted_at":"2026-09-29T21:50:35.054436Z","submitter_name":"HAIXIANG TANG","submitter_email":"hansont2@illinois.edu","task":{"status":"submitted","version":2,"extra_fields":{},"submitted_at":"2026-09-29T21:50:35.054436+00:00","industry_domain":"Hydrology; Environmental Engineering; Scientific Data Quality","task_short_name":"usgs_flood_record_regime_audit_instance_1","operating_system":"Linux","task_description":"Prepare an offline, reproducible flood-record handoff for eight real USGS Kentucky streamgages from the archived professional flood-frequency study through water year 2021. The frozen input contains 841 source-context observation rows, 13 station/regime analysis contexts, selected original PeakFQ specification lines, and the corresponding public summary rows. Read solver/TASK.md for the full visible policies, schemas, scoring formulas, tolerances and completion rules; solver/PROVENANCE.md documents sources and adaptation boundaries.\n\nReconstruct the published configurations and audit eligibility without merging regulated and unregulated contexts. Preserve eight-digit station identifiers, original line references, qualification codes and date representations. Convert known dates to water years and handle unknown month/day precision without inventing dates. Keep historical peaks, explicit intervals, source sentinel bounds and missing discharge distinct. The published systematic record periods are not interchangeable with perception-threshold spans. Respect inclusive segmented periods and the original ordering of overlapping thresholds. Preserve unspecified options rather than inventing defaults.\n\nProduce five artifacts: record_audit.csv, analysis_periods.csv, reconstructed_inputs.json, sensitivity.csv and decision_memo.md. The audit ledger classifies every supplied source row under an explicit priority-ordered contract. The period table reconciles source-row partitions and distinguishes years without nonhistorical records from years with no evidence. The JSON reconstructs threshold/interval provenance, skew choices, global numerical options and reported periods. Sensitivity compares the declared point sample, omission of estimated peaks, and a deliberately period-blind stress test using sample counts, maximum discharge and descriptive log moments. Interpret material differences and remaining domain assumptions in the memo.\n\nDo not replace EMA with sample moments, estimate new design floods, infer undocumented regulation boundaries or thresholds, claim causal dam effects, or call documentary differences proven scientific errors. Full PeakFQ/EMA execution is outside this task. The published frequency values remain visible context and are not the hidden target; the evaluated outputs are the newly derived, traceable qualification, period and configuration audit.\n\nNumerical partial credit is weighted 35% record ledger, 25% period reconciliation, 25% reconstructed configuration and 15% sensitivity. The memo is a required structural gate with no automatic certification of prose quality. Passing requires total score at least 0.85, no missing/unreadable/invalid required artifact, and correct critical historical/interval/daily-mean or imprecise-date records, without fabricated critical records. CSV header/row order and JSON object/list order do not affect evaluation. All definitions and gates are in the solver-visible task.\n\nUse only the frozen inputs. No credentials or live services are needed. These are authentic government hydrologic observations, not synthetic measurements or a machine-learning benchmark. The task is an independent audit adaptation, not an official engineering approval.\n\nProfessional provenance: the underlying USGS study was conducted in cooperation with the Kentucky Transportation Cabinet. Data release: https://doi.org/10.5066/P13HGU8D ; report: https://doi.org/10.3133/sir20265036 . Sources are CC0 with attribution preserved. Task adaptation, reference implementation and validation were AI-assisted; no contributor role in the original government project or human expert sign-off is claimed.","software_licensing":"Free/Open Source","evaluation_criteria":null,"verification_method":"Extract both ZIPs into one directory; they create sibling solver/ and evaluator/ folders. Verify each MANIFEST.json. The solver-visible solver/TASK.md is authoritative for every field, identifier, classification, numeric tolerance and scoring rule.\n\nCompare five required artifacts against evaluator/gold/: record_audit.csv (35%), analysis_periods.csv (25%), reconstructed_inputs.json (25%), sensitivity.csv (15%), plus required decision_memo.md structure. Score keyed records/fields with partial credit, penalizing missing, extra and incorrect values. Pass requires total >=0.85, all declared critical records correct, and no hard failure. The memo must have its required standalone headings and content; its structural gate does not certify prose quality. Missing, unreadable, malformed, empty or duplicate-key required artifacts cause hard failure while other components retain informative partial credit. Row/header order and JSON object/list order are not scored.\n\nRun python evaluator/score.py path/to/solver_outputs and inspect the printed score, passed, hard_failure and critical_records_correct fields, not merely exit status. The grader reads bounded regular files and never executes submitted code.\n\nFor reproducibility, run python evaluator/reference.py --inputs solver --output regenerated_outputs, then score regenerated_outputs. Expected: score=1.0, passed=true, hard_failure=false, critical_records_correct=true; all five regenerated files match evaluator/gold/ byte-for-byte. The separate independent_verify.py checks source and numerical derivations. Run python -m unittest -v test_score from evaluator/ for the mutation fixtures.\n\nThe packaged reference is executed, not a future plan. Actual validation records include 78 passing tests and 9,505 independent checks; timings, code hashes and environment are frozen in evaluator/VALIDATION.json. Packaged file sizes and SHA-256 hashes are listed in solver/MANIFEST.json and evaluator/MANIFEST.json. Local runs used Windows and CPython 3.12.14; Linux is the portable target, not claimed as independently executed. No strongest-model difficulty self-test or human expert validation is claimed.","software_and_version":"CPython 3.12.14, Python standard library only. Linux target; local validation executed on Windows. No R/PeakFQ, licensed software, network, or credentials required.","specific_requirements":null,"expected_completion_time":null},"files":[{"file_id":"e7092a01-02ea-496a-a9ed-86d01054babe","filename":"solver_inputs.zip","category":"input_materials","content_type":"application/x-zip-compressed","size_bytes":25610},{"file_id":"0ef89a4d-deb2-408e-b9cb-a9550be4cf4f","filename":"evaluator_bundle.zip","category":"reference_output","content_type":"application/x-zip-compressed","size_bytes":39601}],"deep_review":{"overall_verdict":"Accept","review_text":"**Overall Verdict:** Accept — The submission is complete, reproducible in substance, and objectively evaluable, with no material blocker identified.\n\n**Completeness Check:**\n\n- Instruction: Complete — Defines the five artifacts, classification priority, date conventions, provenance requirements, and scientific limitations.\n- Inputs: Complete — Contains 841 observations across eight stations and 13 analysis contexts, specifications, summary rows, and the qualification codebook.\n- Toolchain: Complete — The offline Python standard-library workflow is feasible on one ordinary machine without licensed software.\n- Reference output: Complete — All five gold artifacts are present; independent source-to-gold checks found no discrepancies in the ledger, periods, configurations, or sensitivity calculations.\n- Rubric: Complete — Weights, partial-credit formulas, numeric tolerances, structural failures, and critical-record gates are explicit and consistent with the inspected evaluator.\n\n**File and Task Reachability:**\n\nAttachments: `reachable`. End-to-end task: `reachable`.\n\nBoth `solver_inputs.zip` and `evaluator_bundle.zip` were inspected, including every archive member. Their actual sizes match the attachment manifest; archive integrity, safe entries, UTF-8 readability, and all packaged size/SHA-256 declarations were verified.\n\nInspected content includes `TASK.md`, `PROVENANCE.md`, both manifests, source inventories, all observation/specification/summary tables, the codebook, all five gold artifacts, evaluator and reference source, independent-verification source, tests, and validation records. No packaged content remains unreviewed.\n\nThe completion path is self-contained: derive the five artifacts from frozen inputs, then evaluate them against the separately held references. Uploaded programs were not executed; the reported 78 passing tests and 9,505 checks remain contributor-reported validation. Linux execution and external provenance were not independently verified.\n\nLicense keys, accounts, tokens, and other credentials are not required from the contributor. Confidential, proprietary, or internal material may be evaluated as submitted evidence without requiring additional disclosure.\n\n**Special Screening:**\n\nNone. The evidence supports a derived audit of documented government observations, not AI-fabricated measurements or deliberately injected defects. Public provenance alone does not establish a public benchmark; external public availability of the derived answers was not verified.\n\n**Recommended Changes:**\n\nThe source traceability, explicit scientific boundaries, separated references, and defensive grading contract are substantive strengths. No material changes are required. Optionally qualify the cross-platform byte-for-byte claim or pin output newlines: the JSON and memo contain Windows line endings, while Linux regeneration would normally differ in those bytes without affecting evaluation.","reviewed_at":"2026-10-14T19:47:54.308388Z"},"claim":{"id":"e2f4cf04-b990-4add-89b8-eed7a11c7fd8","submission_id":"c4e3b0be-7a98-4267-b82a-7519be2d267c","submission_revision_id":"6d1c1bd8-af1c-4908-9af3-442f6b92d962","builder":"builder-01","status":"active","claimed_at":"2026-10-15T15:34:37.083309Z","delivered_at":null,"ended_at":null,"delivery_note":null,"end_reason":null}},"revision_is_current":true}