zECS assembler: ZECS002 / ZECSNC refused whole, ZECS001's CALL ZUIDSTCK; frontier (Gap Atlas) #5859

Owner decision 2026-10-21: frontier. The zECS assembler programs Source/ZECS002.asm and Source/ZECSNC.asm are refused whole ("refused: CICS Assembler"): the det translator is a COBOL translator. They count in the 129-program burned-estate ladder denominator (both at L0) but are not core work toward this finish line.

What it would take (mapped, not built): a CICS Assembler (HLASM + EXEC CICS) front end or a per-program port path, an oracle for assembler, and z/OS-faithful register / storage semantics.

Owner decision 2026-10-21: ZECS001 is frontier too. Source/ZECS001.cbl translates 560 of 561 statements; its one hole is CALL ZUIDSTCK, an assembler store-clock wrapper whose source the estate does not hold, so the 16 bytes it returns in THE-TOD are not a stated fact (oracle_assumptions.md X47). Closing it needs either the assembler path below or a ZUIDSTCK stand-in with a modelled runtime on both sides (system-copybook rule); neither is core work for this finish line.

• (frontier) Decide whether the ladder report shows refused-whole assembler programs separately from the COBOL denominator (progress page Generated progress page + README numbers that can't drift: burned-estate ladder, per-tool %, core burn-down with ETA, Gap Atlas #5280)
• (frontier) Assembler translation path, if ever pursued
• (frontier) ZECS001: CALL ZUIDSTCK (STCK wrapper): a stand-in with a modelled runtime, or the assembler path

Scope: frontier (Gap Atlas). Filed from the 2026-10-21 translation-blocker audit.