All checks were successful
CI / lint (push) Successful in 35s
Deploy / notebooks (push) Has been skipped
Deploy / zotero (push) Has been skipped
Deploy / docs (push) Has been skipped
Deploy / api (push) Has been skipped
Deploy / mc (push) Has been skipped
Deploy / report (push) Successful in 16s
CI / test (push) Successful in 13m3s
The self-consistency check previously skipped in headless CI (no aco DuckDB), so CI only guarded that the CFs were *unchanged*, not that they were *correct*. Add a committed slice of 30 real Addendum B rows (fixtures/addendum_b_sample.csv — real published payment_rate, weights ~0.05 to ~700, CY2021-CY2026) and reconcile OppsPricer against it in an in-memory DuckDB. This runs in CI with no external data, and because the rows are real, a wrong conversion factor makes weight × CF diverge from the published rate and fails the build. Verified fail-closed: reverting the CY2026 CF to the old wrong value fails the golden test, this fixture test, and the full-DB check together.