Original reusable-operations-v0 result. Read record/REPORT.md and MANUAL_REVIEW.md; the original saved selection, unknown data and expenses remain unchanged. The prior source archive is included byte-for-byte. No raw task content, prompts, provider payloads, grader source, or model-authored library programs are in this result ZIP. run/libraries-frozen.json preserves collection digests/counts/provenance; those do not validate semantic portability. Original complete results can be displayed without benchmark access or inference: python -m scripts.report_reusable_operations --study record --output reproduced-tables.md. Compare with record/RESULT_TABLES.md. This reproduces saved numerical tables, not a fresh stochastic benchmark score.