The same task was run independently twice and both results edit the header of graphs/backlog-batch.yaml. Branch formalise-batch-lane (PR #237, unmerged; changelog commit 82146c6) changes five files including internal/graph/shipped_graphs_test.go. Branch lane-b2b (commits 48e32d2, 9374112, 264d464, d97c652) ships no test at all and adds docs/measurements/0034b-independent-lane-failure-predicate.md instead. The consequence is that any citation of graphs/backlog-batch.yaml: resolves differently per branch. Measured 2026-08-22 with git -C /private/tmp/w-b2b grep -n '^on_fail' -- graphs/backlog-batch.yaml: formalise-batch-lane puts the declaration at :152, lane-b2b at :138 (it was :133 before d97c652 added the four-line verdict header on the same branch). What is needed is a decision about which branch survives, and then a re-check of every graphs/backlog-batch.yaml address recorded against the loser.
The same task was run independently twice and both results edit the header of graphs/backlog-batch.yaml. Branch formalise-batch-lane (PR #237, unmerged; changelog commit 82146c6) changes five files including internal/graph/shipped_graphs_test.go. Branch lane-b2b (commits 48e32d2, 9374112, 264d464, d97c652) ships no test at all and adds docs/measurements/0034b-independent-lane-failure-predicate.md instead. The consequence is that any citation of graphs/backlog-batch.yaml: resolves differently per branch. Measured 2026-08-22 with git -C /private/tmp/w-b2b grep -n '^on_fail' -- graphs/backlog-batch.yaml: formalise-batch-lane puts the declaration at :152, lane-b2b at :138 (it was :133 before d97c652 added the four-line verdict header on the same branch). What is needed is a decision about which branch survives, and then a re-check of every graphs/backlog-batch.yaml address recorded against the loser.