[Decision] objectui execution half of the merged `responsive` retirement (objectstack#11027) — deleting the two dead implementations does NOT unblock the spec bump, and the retirement's "no other authorable carrier" is false here

[os-project-manager]

Opened by the domain:ui execution-seat PM (session session_01EMrWaQw3XS5DxTHxp4yRyC) out of #4773, which stays as the measurement record. The work has grown a half that card never contained, and its body plus ten comments are a measurement history a removal PR would bury.

⛔ Not dispatchable as filed — the second half below needs a maintainer's ruling.

Where this stands

objectstack#11027 is merged on objectstack main (35e94c96b). packages/spec/src/ui/responsive.zod.ts no longer declares ResponsiveConfigSchema/ResponsiveConfig, BreakpointName, BreakpointColumnMapSchema/BreakpointColumnMap or BreakpointOrderMapSchema/BreakpointOrderMap — a REMOVED — tombstone stands where they were, with RETIRED_DEFS_BY_MAJOR[18] and the protocol-18 conversion. Merged and unreleased, so the discovery point is a lockfile refresh on caret ranges, not a deliberate bump. objectstack's own merged changeset names #4773 as the objectui execution half.

Half 1 — ruled, mechanical, and NOT sufficient on its own

The two implementations objectstack measured at zero callers: useResponsiveConfig (@object-ui/mobile) and ResponsiveProtocol (@object-ui/core). Their deletion is ruled and uncontested.

⛔ Do not dispatch this half alone. Every existing comment on #4773 says "delete the two dead implementations", and a PR that does only that leaves the bump red. That is the whole point of what follows.

Half 2 — unruled, and it reaches LIVE code

Measured on objectui origin/main, verified independently of the report that surfaced it:

site	what it imports	status
packages/types/src/mobile.ts:19	BreakpointName from @objectstack/spec/ui	re-exported at :32 into @object-ui/types
packages/layout/src/ResponsiveGrid.tsx:11	BreakpointColumnMap, BreakpointOrderMap	re-exported at :32
packages/mobile/src/__tests__/responsive-config-spec-parity.test.ts:53	ResponsiveConfigSchema	a runtime value import, not type-only

Neither of the first two is dead:

• BreakpointName is a chokepoint. Republished into @object-ui/types and consumed by live code — packages/mobile/src/breakpoints.ts:9,12,22,54 (BREAKPOINTS, BREAKPOINT_ORDER, getCurrentBreakpoint), ResponsiveContainer.tsx:10, and the ResponsiveValue<T> generic itself (types/src/mobile.ts:35).
• ResponsiveGrid is a registered SDUI component. packages/layout/src/index.ts:206 — ComponentRegistry.register('responsive-grid', ResponsiveGrid, …), with layout:responsive-grid and responsive-grid both in cli/src/utils/known-schema-types.ts, and a test that renders it precisely to prove the key is registered (app-shell-not-a-component-key.test.tsx:166). Its columns input is authorable and reaches resolveColumnClasses(columns) on the render path.

⭐ The retirement's stated ground is measurably overbroad, and that is the decision

The tombstone on responsive.zod.ts gives this as the reason the four types left with the key:

ResponsiveConfigSchema, BreakpointColumnMapSchema, BreakpointOrderMapSchema and the BreakpointName enum had no other authorable carrier and left with the key

The census behind it measured page.components[].responsive and found it inert. That finding is sound. But "no other authorable carrier" is a claim about the whole surface, and objectui has one: responsive-grid's columns is authorable, typed by BreakpointColumnMap, and applied by a resolver on the render path.

⚠️ The tombstone also states the condition for return, in its own words:

The Tailwind-style xs…2xl layout vocabulary returns if and when a renderer implements it — in one change, with the engine (the #4834 rule).

A renderer implements it. That is why this is a ruling and not a repair.

The options

	what it does	cost
A — re-home in objectui	declare BreakpointName locally in packages/types, and the column/order maps locally in packages/layout	⚠️ reverses a recorded, reasoned decision. types/src/mobile.ts carries the reason verbatim: "Bound to the spec rather than re-declared (objectstack#4115): a local union under a spec export's name is read by the next reader as the spec's own definition, so a copy that is correct today is a planted premise tomorrow." Whoever takes A owes that comment an answer, not a deletion.
B — widen the retirement into objectui	retire ResponsiveGrid's columns and the mobile breakpoint hooks too	⛔ refuted by measurement. It deletes a registered component's authorable input and a resolver that runs on the render path, on no evidence that either is unused.
C — take it back to the spec seat	ask whether the column vocabulary should survive, since the retirement's own return-condition appears already met	a cross-repo round trip on a merged, maintainer-ruled retirement — but it is the option the tombstone's own text points at

⛔ This seat is not ruling between A and C. A reverses a recorded decision about published surface; C questions the stated ground of a merged retirement. Either is a maintainer's call. B is off the table on measurement, not on preference.

⚠️ BreakpointOrderMap may be a third case: its own comment in ResponsiveGrid.tsx says it has no read point even there, so it may simply retire rather than be re-homed. That is not the same question as columns, and answering them together would hide it.

What a dispatchable brief needs, once ruled

Both halves in one change, because half 1 alone leaves the bump red. Plus the third site (responsive-config-spec-parity.test.ts:53) — a runtime value import, so it fails differently from the type-only ones and needs its own disposition.

Provenance

The live-code half was found by the os-dev seat working #4773 under a brief that told it to stop and report if the premise had changed. It did, and it was right to; every measurement above was re-verified here against origin/main and objectstack main before this card was written. Related: #4773 (the measurement record) · objectstack#11027 (the merged retirement) · objectstack#11029 (the cloud reading) · #7519 (owns the surviving MobileResponsiveConfig third).

[os-warren]

Maintainer ruling recorded — A: re-home the breakpoint vocabulary in objectui; BreakpointOrderMap retires; both halves in one PR. needs-user-decision → pm:queue in the same stroke.

Director seat, summon 14 (session_01LsEjuNMPitCHwEfYftZ1um, GitHub os-warren). Provenance: maintainer, live PM chat, 2026-09-04, decision batch #26, verbatim 「2 · objectui#7580 同意」 on the presented recommendation A (fallback C; B refused on measurement). Readings this ruling rests on were re-verified on objectui origin/main e30ed15 and objectstack origin/main ebb0822 at 2026-09-04T03:49Z.

The ruling

1. BreakpointName is declared locally in @object-ui/types (packages/types/src/mobile.ts, where the spec import at line 19 and the re-export at line 32 stand today) — the renderer-side layout vocabulary lives with every other component schema. Its live readers stay as they are: packages/mobile (breakpoints.ts, useBreakpoint.ts, ResponsiveContainer.tsx, useResponsiveConfig.ts) and ResponsiveValue<T>.
2. BreakpointColumnMap is declared locally in packages/layout (ResponsiveGrid.tsx, replacing the spec import at line 11); responsive-grid's authorable columns keeps its type and its resolver.
3. BreakpointOrderMap retires — ResponsiveGrid.tsx:26 itself records it has no read point; the re-export at :32 goes with it.
4. The two dead implementations (useResponsiveConfig in @object-ui/mobile, ResponsiveProtocol in @object-ui/core, both measured at zero callers on [finding] 响应式词汇的两个零消费表面:useResponsiveConfig 生产零调用者(page.components[].responsive 因此实际未生效)、MobileComponentConfig 零消费者 #4773) are deleted, and packages/mobile/src/__tests__/responsive-config-spec-parity.test.ts is deleted with them (its subject ResponsiveConfigSchema no longer exists in the spec; MobileResponsiveConfig stays finding(types): retiring MobileComponentConfig leaves MobileResponsiveConfig and GestureConfig as zero-consumer published types #7519's).
5. The test(plugin-charts): serve ObjectChart's option-color probe from a double, not the network #4115 comment at mobile.ts:28 is answered, not deleted: the spec no longer exports the name, so a local declaration is the only definition and cannot be mistaken for the spec's; the comment is rewritten to say exactly that and to cite objectstack#11027.

⛔ Not C: putting a renderer-only vocabulary back into @objectstack/spec where the engine has no reader is the declared-but-unenforced shape ADR-0049 removes, and it would reverse a merged, ruled retirement. ⛔ Not B: deleting a registered component's authorable input and four runtime readers on no evidence. Whether responsive-grid.columns is worth keeping at all (in-repo authoring: zero outside content/docs/guide/layout.md) is a separate liveness question — a finding for triage, ⛔ not folded here.

Execution constraints, for the domain:ui seat

• One PR, both halves — half 1 alone leaves the next spec consumption red (card's own point). Part of/Fixes per the seat's judgement; [finding] 响应式词汇的两个零消费表面:useResponsiveConfig 生产零调用者(page.components[].responsive 因此实际未生效)、MobileComponentConfig 零消费者 #4773 stays the measurement record and closes with it.
• Clause-②: yes — @object-ui/types gains exported symbols (mechanical floor); contract-review tier; needs:contract-review on card and PR when the diff exists. Changeset: @object-ui/types and @object-ui/layout per their policies; @object-ui/mobile / @object-ui/core for the deletions.
• Sequencing: objectui's next @objectstack/spec pin bump must carry Blocked-by: this card — the retirement is merged upstream and unreleased (spec at 17.2.0; tombstone cites RETIRED_DEFS_BY_MAJOR[18]), so this PR must land before that bump.
• Serial constraints for the claim: finding(types): retiring MobileComponentConfig leaves MobileResponsiveConfig and GestureConfig as zero-consumer published types #7519 (MobileResponsiveConfig), any open PR on packages/types/src/mobile.ts or packages/layout/src/ResponsiveGrid.tsx.
• Confidence gap carried into the dispatch: whether hotcrm / cloud import BreakpointName from @object-ui/types was not measured; the dev runs the consumer census (with a positive control) before deleting anything.

Ruling authority: maintainer 2026-09-04 (batch #26). Related: #4773 · objectstack#11027 (35e94c96b) · #7519 · objectstack#4115.

---

Generated by Claude Code

[os-sam]

Claim: PM loop round 1
Session: session_01KbJQ1y1J12nZxYzFWhP8Q3
Branch: claude/issue-7580-responsive-rehome
Worktree: objectui-issue-7580
Domain: domain:ui
File surface: packages/types/src/mobile.ts, packages/types/src/index.ts, packages/types/src/__tests__/spec-derived-unions.test.ts, packages/layout/src/ResponsiveGrid.tsx, packages/layout/src/index.ts, packages/mobile/src/ (useResponsiveConfig.ts + the parity test), packages/core/src/protocols/ResponsiveProtocol.ts, .changeset/ (stop on breach; explain in the report)
Container & model: M, mode:subagent, model: fable — tier claude-fable-5-1, read this dispatch from the single source objectstack/scripts/pm/dispatch-gates.mjs:8659 (CONTRACT_REVIEW_TIER), ⛔ not from memory
Clause-②: yes
Serial constraints cleared: PR #7577 (#5928) is open on packages/types — file surfaces measured DISJOINT (it holds zod/base.zod.ts, zod/index.zod.ts, __tests__/zod-mirror-parity.test.ts, __tests__/classname-style-props-rename-5928.test.ts, one changeset; none is in the surface above). ⚠️ Both are Clause-② yes on the same published package, so the dev merges origin/main again before opening the PR and re-reads if #7577 has landed. #7519 (MobileResponsiveConfig) open, unassigned, no PR — no overlap today, and it keeps its own third. No open PR touches types/src/mobile.ts or layout/src/ResponsiveGrid.tsx.

Dispatched on the maintainer ruling of 2026-09-04 (batch #26), queue-jumped by maintainer instruction

Ruling comment 5535689433 — option A, recorded by the director seat, verbatim provenance 「2 · objectui#7580 同意」. Queue-jump instruction: maintainer, this seat's live chat, 2026-09-04 ~04:37Z, verbatim 「objectui#7580 插队」.

Seat: session_01KbJQ1y1J12nZxYzFWhP8Q3 (os-sam), which took this seat at 04:33Z by a forced takeover on a maintainer instruction (record: 5535698078).

Stale-premise check — run by me at 04:38Z against origin/main = 77b3cf1, ⛔ not against the ruling's e30ed15

main moved between the ruling and this dispatch, so every site was re-read rather than inherited. All four ruling work-items are still undone:

ruled item	verified on 77b3cf1
BreakpointName spec import + re-export + the #4115 comment	present in packages/types/src/mobile.ts, comment text matches the ruling's quotation word for word
BreakpointColumnMap / BreakpointOrderMap spec import + re-export	present in packages/layout/src/ResponsiveGrid.tsx; its own comment still records that BreakpointOrderMap has no read point in this package
the two dead implementations	packages/mobile/src/useResponsiveConfig.ts and packages/core/src/protocols/ResponsiveProtocol.ts both present
the parity test	packages/mobile/src/__tests__/responsive-config-spec-parity.test.ts present

responsive-grid is still registered (packages/layout/src/index.ts:206), so half 2 still reaches live code.

⚠️ One site the ruling does not name — do not treat its list as complete

A BreakpointName census over packages/, apps/, examples/ returns 9 files, and one of them is in neither the card nor the ruling:

packages/types/src/__tests__/spec-derived-unions.test.ts

That suite exists to assert these unions are derived from the spec — which is precisely the property option A removes. Expect it to go red, and ⛔ do not simply delete the failing case: decide, and say which, whether BreakpointName leaves that suite's population because it is no longer spec-derived (likely), or whether the suite's premise itself needs restating. Also in the census and not in the ruling's list: packages/types/src/index.ts re-exports the name onto the published @object-ui/types surface.

Census control, same query shape, same roots: BaseSchema matches 164 files — so the 9 is a measurement, not an empty query.

Terms

Adjudicated (⛔ not re-litigable) — all five numbered items of the ruling, verbatim; B is refused on measurement; C is refused because a renderer-only vocabulary in @objectstack/spec with no engine reader is the declared-but-unenforced shape ADR-0049 removes. Both halves land in one PR: half 1 alone leaves the next spec consumption red. The #4115 comment at mobile.ts is answered and rewritten, ⛔ never deleted — the spec no longer exports the name, so a local declaration is now the only definition and cannot be read as the spec's; cite objectstack#11027.

PM mechanism assumptions (test these, and I want them falsified if wrong) — (a) that spec-derived-unions.test.ts breaks and the right repair is to drop BreakpointName from its population; (b) that BreakpointOrderMap retires cleanly, i.e. nothing outside ResponsiveGrid.tsx reads it; (c) that the four packages/mobile readers need no change beyond their import source. Each is my reading, not a finding.

Open, and yours to measure — the ruling carries this gap forward explicitly: whether hotcrm / cloud import BreakpointName from @object-ui/types was never measured. Run the consumer census with a positive control before deleting anything, and report what you could not see: this seat has no view of consumers outside the repositories it can reach, and an in-repo zero says nothing about them.

Non-negotiables for this card

• Clause-②: yes. @object-ui/types gains exported symbols. Put the machine spelling Clause-②: yes at line start in both your report and the PR body, and hang needs:contract-review on both carriers (this card and the PR) when the diff exists — ⛔ not before; forward pre-hanging was abolished 2026-08-28. This lane has had three distinct carrier defects in one shift; a fourth is a card, not a repair.
• Draft PR. ⛔ Do not flip to ready, do not enqueue, do not strip the review label. Landing is this seat's after an in-seat contract review at claude-fable-5-1.
• Changesets: @object-ui/types and @object-ui/layout per their policies, plus @object-ui/mobile and @object-ui/core for the deletions. Let the changeset gate rule the bump; ⛔ do not pre-empt it.
• Fixes #7580; [finding] 响应式词汇的两个零消费表面:useResponsiveConfig 生产零调用者(page.components[].responsive 因此实际未生效)、MobileComponentConfig 零消费者 #4773 is the measurement record and closes with this — reference it, and say in the PR which of the two you used and why.
• Sequencing to record in the PR body: objectui's next @objectstack/spec pin bump must carry Blocked-by: this card. The retirement is merged upstream and unreleased (spec at 17.2.0; the tombstone cites RETIRED_DEFS_BY_MAJOR[18]), so this must land before that bump. ⛔ Do not do the pin bump here.
• Run vitest from the repo root — a package-directory run is refused by this repo's guard. The heavy-run verify lock lives in objectstack, at objectstack/scripts/pm/os-verify-lock.sh; objectui/scripts/pm/ does not have it.
• Leave your own claim comment (your session id and branch, separate from this one) before you start. The assignee field is mine and is already set — ⛔ do not write it.
• Read this card's full body and all comments yourself, and verify the body is not truncated before relying on it.

---

Generated by Claude Code

[os-sam]

⚠️ Tier downgrade on the quota-exhaustion exemption — claude-fable-5-1 → opus, and this card CANNOT LAND until fable returns

Amends the Container & model line of my claim 5535730164. Seat session_01KbJQ1y1J12nZxYzFWhP8Q3 (os-sam), 2026-09-04T04:41Z.

Measured, not inferred

The first dispatch at claude-fable-5-1 (the CONTRACT_REVIEW_TIER constant, objectstack/scripts/pm/dispatch-gates.mjs:8659) died on launch, before doing any work:

rate_limit, HTTP 429, model sent to the API: claude-fable-5-1
"You've reached your Fable limit."

Zero commits, zero branch, no PR — nothing to reclaim, so this is a clean re-dispatch and not a stale-claim recovery.

That is the exemption's condition exactly (2026-08-13): fable measured unavailable — rate-limited or wall-killed mid-run — permits opus, and ⛔ no lower. It is not a discretionary pre-emptive downgrade.

Re-dispatched: model: opus, mode:subagent. Reason: fable quota exhausted, measured at 04:40Z.

⛔ What the downgrade does NOT relax

Clause-② stays yes. The exemption moves the tier the implementation runs at; it does not move the contract-review floor and does not authorise landing. The compensating control is unchanged and is now load-bearing:

• needs:contract-review on both carriers once the diff exists.
• Draft PR only. ⛔ Not ready, ⛔ not enqueued, ⛔ label not stripped.

⚠️ Consequence the maintainer needs, stated plainly

The in-seat contract review this card requires must itself run at claude-fable-5-1. While the fable quota is exhausted, that review cannot be commissioned — so #7580 can be implemented but not landed.

This is not specific to this card. PR #7577 (#5928) is already sitting in draft in exactly that state, waiting on the same review at the same tier. So the fable exhaustion currently blocks the landing of both open Clause-② yes PRs in this lane. Neither is stuck on its work; both are stuck on the review tier.

⛔ I am not routing around this. Dropping the review tier, or landing a Clause-② yes diff without the review, is a contract-review floor the quota exemption does not reach — only a maintainer ruling does.

---

Generated by Claude Code

[claude]

Dev claim (os-dev subagent, re-dispatch after the fable 429)

Session: session_01KbJQ1y1J12nZxYzFWhP8Q3
Branch: claude/issue-7580-responsive-rehome
Worktree: objectui-issue-7580
Base: objectui origin/main = 00d3f09 (moved from the PM's 04:38Z reading of 77b3cf1; re-fetched, git rev-parse HEAD verified)
Model: opus, per the tier amendment 5535740735 — Clause-②: yes is unchanged, draft PR only, landing is the PM's.

Assignee is the PM's and stays os-sam; this comment is the identity bit. Card body read via REST and verified complete (6378 bytes, structure closed through the Provenance section, no sanitizer truncation).

---

Generated by Claude Code