Repository navigation
feat(review): add a per-heading act-on cap to fanout's report #12503
Description
Parent: #7584
Problem
#7584's decisions read "act-on cap per heading (setting from the Q160 eval)" (D101). #12151's run measured, per heading, the largest position at which the planted defect appears: under lens (the arm the verdict keeps as default) it is 1 under ### Spec and 1 under ### Standards, across 36 VALID runs; model gives the same. The bar is the smallest cap that keeps every planted defect inside "act on".
Done when
- A setting sets the act-on cap per heading, externalized like
fanout_diversity, with its default taken from test(review): add fanout-diversity fixtures that separate the arms #12505's multi-defect positions, not from test(review): run the fanout reviewer-diversity eval and post its probe result #12151 alone. - Fanout's report marks the rows inside the cap as "act on" and the rest as not, without dropping any row.
- A test or eval case shows the cap applied per heading.
Evidence: #12151's probe result. Both headings have evidence; neither is none. The fixtures are at ceiling (both arms 24/24), so harder fixtures (filed alongside) may move this default.
The #12151 position of 1 per heading is a floor, not a default: each fixture plants one defect in a small diff, so position 1 is close to guaranteed, and a change with two serious defects under one heading would have the second left unmarked.
Blocked by: #12505
Activity
github-actions commentedon Nov 2, 2026
Proposed work class: work-class: scoped
Basis: automated intake triage; the lanes App applied this class.
Size: Small
Complexity: Low
Uncertainty: Known
Flags: none
Blocked by: #12505
Understood: add a per-heading "act on" cap setting to fanout's report, externalized like
fanout_diversity. Rows inside the cap are marked "act on" and the rest are not, with no row dropped. A test or eval case should show the cap applied per heading. The default comes from #12505's multi-defect positions, so this waits on that issue. No duplicates found.Automated triage, applied by the lanes App.
github-actions commentedon Nov 2, 2026
The triage sweep dispatched intake triage for this issue (attempt 1 of 2).
melodic-automation-lanes commentedon Nov 2, 2026
Not admitted to the issue lane yet: blocker #12505 is not closed by a listed human or a merge.
github-actions commentedon Nov 4, 2026
github-actions commentedon Nov 4, 2026