The reading assembler test's tree walk lstats every path it enumerates without tolerating ENOENT, so a transient file under .git - git's background maintenance.lock - that exists at enumeration and is gone by the stat fails the test; it flaked TestAssembleDryRunWritesNothing on the macOS CI leg and is a time-of-check race any concurrent git activity can trigger

The reading assembler test's tree walk lstats every path it enumerates without tolerating ENOENT, so a transient file under .git - git's background maintenance.lock - that exists at enumeration and is gone by the stat fails the test; it flaked TestAssembleDryRunWritesNothing on the macOS CI leg and is a time-of-check race any concurrent git activity can trigger

Grounds

• pursued: a snapshot of the worktree has no business walking .git at all, and tolerating the error would have kept the race while hiding it. What would show this wrong is a dry run that writes into .git, which nothing in the assembler can do.