[Hafeok]

What this changes, and why

applications/sdlc/sdlc.md §19 gets two new open items, as Emil asked in Round 31. Neither rules on its substance. One commit, b79d326, off main at a53d02d. It is kept out of #65.

• O-8: an act that accepts the realisation plan.
  • The gap. o1b plans the realisation on the issue, but no act accepts the plan before work starts. o2b reads the plan, and o10 accepts only the candidate, after the work.
  • Where one exists. product-cli's issue sessions have one (product-cli ADR-0014 D5, ADR-0014: issue sessions run as a Claude Code routine product-cli#171, Proposed there).
  • The proposal, unruled: a plan-acceptance act between o1b and o4, by the How owner or a holder of its grant. Its verdict is in o4's ground reading.
  • The row says where it would sit:
    • in the record (§13): its address, lc:after, lc:readsIssuePosition, its verdict and its waits. o4's lc:after names it, and the bootstrap carries the accepted plan's position;
    • in the meters (§14, §18): a new human-wait cost line, read on the o5 ↔ o6 hotspot's How class. Placed after o2b, it falls inside flow time. In either place, it falls outside DORA's lead time.
• O-9: reconciling product-cli's issue sessions with the SDLC.
  • The two specifications side by side: label, plan approval, landing and record, each in product-cli's form and in the SDLC's.
  • product-cli is the natural first adopter. It either adopts the SDLC or records its departures.
  • Closes it: Emil, then product-cli.
• meta/sessions/2026-11-07-sdlc-and-pdlc/rulings.md gains Round 31, verbatim, after Round 30. It has a "How it governs" note in the record's usual form. The blockquote round-trips to the message byte for byte.

Rulings and decisions

• Round 31 is recorded. Nothing is ruled.
• No decision is filed or amended. DDD-dec-115 stays as landed, since nothing is ruled.
• Basis: DDD-dec-115.

Predicted, then measured

• Predicted: text only in sdlc.md §19 and rulings.md, so scripts/ci.py and both graph checkers pass as on main.

• Measured locally: scripts/ci.py ends all gates passed. applications/pdlc/graph/check.py and applications/sdlc/acts/check.py both exit 0.

• Not predicted: the first push, 681704c, failed CI's issue-refs gate. It cited only product-cli issues, and the gate needs a Refs #N to an issue in this repository.

  I amended the commit to add Refs #61, the session's issue ("Every commit of the session carries Refs to this issue"). canon issue-refs --basis then passed locally, and the validate run on b79d326 is green.

What was found

• Neither item was already covered.

  • Rounds 27 to 30 close O-4 to O-7 and the pdlc.md items.
  • sdlc.md has no act that accepts o1b's plan.
  • sdlc.md never mentions product-cli's issue sessions.

• Numbering (UNVERIFIED — Emil review). O-8 and O-9 are §19's next free numbers. pdlc.md §15 has its own O-8 (the What elements) and O-9 ("flow time", closed by Round 27), and Round 27's text says "open item O-9" for the latter.

  The two lists are separate, and Round 31's note says each item is cited with its document. Separately, O-8 here is not the retired act o8.

• One fact added to O-9 (UNVERIFIED — Emil review). It is not in Round 31's text. Under product-cli ADR-0014 D5, the dispatched run writes the plan itself. So o1b is occupied after o2b and o3, while §5 plans before allocation.

  This is drawn from ADR-0014's text and is not a ruling. Strike it if you want O-9 limited to Round 31's list.

• One question O-8 leaves open: whether a model may occupy the plan-acceptance act, as lc:ModelOccurrenceShape forbids at o10.

What is left

• Emil closes O-8, and O-9 with product-cli.
• In product-cli, the step-1 PR for product-cli#169 names O-9 as an open item and does not reconcile it.
• Not done here: landing.md is not updated. Round 31 did not ask for it.

---

• python3 scripts/ci.py passes (every gate: claims, decisions, core order and pins, the What, the How examples, fixtures)
• Every commit that changes canon carries a Basis: line
• Decision ids checked across every branch of both repositories. Not applicable: no decision is filed.
• A session record exists under meta/sessions/: Round 31 is recorded in the 2026-11-07 session's rulings.md, since §19 belongs to it.
• No customer-identifying material anywhere in the diff

Refs #61, Refs mindovermachine-dev/product-cli#169, Refs mindovermachine-dev/product-cli#171

---

Generated by Claude Code

[Hafeok]

approve 80e9088