# E6BR-5 target spec (exact golden finalized at S3.9)

Unlike the harvested `08xx` resolver-target cases — whose exact target bytes were
produced by the E6b branch and are frozen in `*.exit/*.stdout/*.stderr` here —
**E6BR-5 has no oracle that produced exact bytes today** (it is the open hole that
paused E6b; no implementation ever emitted its diagnosed form). Its target is
therefore recorded as a **specification**, not frozen bytes:

- **exit:** `1` (see `e6br5-nested-pack-source-ambiguous.exit`).
- **stderr:** at least one `error: type 'Box' is ambiguous: it is declared in
  multiple flat-imported modules; qualify the reference or remove the duplicate
  import`, pointing at the nested `*Box` leaf inside
  `Closure(Closure(*Box, ..$inner) -> $IR, ..$args) -> $R`.
- **stdout:** empty (the build fails before `main` runs).

The byte-exact `stdout`/`stderr` goldens are generated by the Fork C resolver when
this case **flips to active + green at S3.9** (`run-confirm-r3`: "E6BR-5
nested-pattern case diagnosed"). At that point this `.target.md` is replaced by
real `e6br5-nested-pack-source-ambiguous.stdout` / `.stderr` files and the case
moves to an active marker, exactly like the `08xx` flips.

This case is the re-filed E6BR-5 regression — **explicitly NOT an E6b attempt-6.**