chore(objectui): bump the console pin to 62597c588072 (carries the fail-closed quick_actions requiredPermissions fix) with 46 re-measured citations, the FormField.span describe re-point and the lockstep re-record

[objectstack-fleet]

Fixes #19503

Clause-②: no

Moves the console pin to objectui main 62597c588072. That tip carries objectui#10153, the fail-closed record:quick_actions.requiredPermissions fix. The PR also carries the bump's two required follow-ups, folded in under the PM ruling on #19503 (comment 5791879469, option A): the sdui-parser lockstep re-record, and a re-MEASURE of the packages/spec/src/** citations that check:objectui-pin-citations names. The citation edits touch comments and migration-rationale prose only; there are no schema or export changes. The one published sentence that moves is the FormField.span .describe() pin citation, authorised by the second ruling (comment 5793509969) and carried with .changeset/19503-pin-bump-describe-corrections.md (@objectstack/spec: patch) plus a regenerated content/docs/references/ui/view.mdx. Clause-②: no, because every authored document is accepted and refused exactly as before.

check:objectui-pin-citations exits 0 on the head: "47 asserting objectui pin citation(s) match .objectui-sha (62597c588) … 7 anchor content assertion(s) verified".

Ancestry reading (fix → new pin: AHEAD)

Measured in a FULL (non-shallow) clone of objectstack-ai/objectui. git rev-parse --is-shallow-repository returns false, so an exit 0 is self-proving:

command	exit	reading
git merge-base --is-ancestor 28b065800c7d1b6069c06aa906d0716377b5e281 62597c588072636e9c30ea35b3d89b1e46fd765d	0	the fix is IN the new pin
git rev-list --count 28b065800c7d..62597c588	33	the new pin is 33 commits ahead of the fix
git merge-base --is-ancestor 87af769e9a3e 28b065800c7d	0	the old pin was 4 behind the fix (the card's REST compare: behind_by 4)

The bump's own changeset lists the fix: fix(plugin-detail): record:quick_actions.requiredPermissions is an ADR-0066 capability set, read fail-closed ... (objectui 28b065800).

What moved, and how each artefact was produced

artefact	producer
.objectui-sha + .changeset/console-62597c588072.md	bash scripts/bump-objectui.sh --no-commit 62597c588072… exit 0 (27 releasing changesets, 0 breaking, @objectstack/console: minor)
scripts/sdui-manifest.record.json	node scripts/gen-sdui-manifest-node.mjs --objectui-version 17.6.0 exit 0. sdui.manifest.json came out BYTE-IDENTICAL (sha256 49211fee7792, 57 components), because packages/core ships 17.6.0 at both pins; only the record moves
packages/sdui-parser/objectui-lockstep.json	pnpm gen:sdui-lockstep with OBJECTUI_ROOT at a checkout of 62597c588, exit 0. Diff: 3+/3-, only rev, revDate and recordedAgainstPin. Grammar blob 0131f27cf86d unchanged, 214 lines, all 26 diagnostic codes agree. check:sdui-lockstep exit 0: "byte-identical to objectui@62597c588072". No parser port is owed
packages/spec/src/migrations/registry.ts	pnpm --filter @objectstack/spec gen:migration-registry, from the six edited entries
the console	bash scripts/build-console.sh exit 0: "Console dist ready (60524 KB) from objectui@62597c588072", bundle canary present. check:console-sha exit 0

Citation re-measure: method

• Read-point records. For every cited objectui file I ran git diff --quiet 87af769e9a3e 62597c588 -- FILE.
  • Unchanged file: the anchor holds by identity. The prose names that, keeps the earlier hop's history, and says where the last content re-read happened.
  • Changed file: each anchor was mapped through the git diff -U0 hunks and its text compared at both pins. It was then re-read in the new tree.
• Corpus counts. These are the six migration entries plus registry.ts. I re-counted with git grep -o -F TOKEN SHA | wc -l. The method was calibrated first: at 87af769e9a3e it reproduces every recorded count exactly (8228 files, useState 2383, timeout 1075, objectstack 12966, @objectstack/spec 4997, window 3464, interval 170, tenant 976, …).
  • Every dark token still counts 0 at 62597c588: the def exports and key names each entry names, 95 tokens in all.
  • The Span / SpanSchema collisions are unchanged at 486 / 53.
  • Only the corpus size (8303) and the lit controls move.

Re-read sites (all 46 the gate named)

site(s)	objectui file(s)	reading at 62597c588
6 migration entries + their 6 registry.ts copies	whole corpus	re-counted: dark tokens still 0. The kernel-health-check… entry keeps its 87af769e9 lit-control clause verbatim (now dated) and adds the new counts after it. check:future-spec-major holds a measurement exemption to that exact sentence, and the new figure "@objectstack/spec 5043" would read as a version
component.test.ts accordion/tabs icon; component.zod.ts PageTabs / PageAccordion icon	containers.tsx (changed, 4c6f549ef)	re-read: 853-859, 912, 1069-1075 and 1116 did NOT move, text byte-identical
component.test.ts button icon; component.zod.ts element:button icon	resolve-icon.ts, button.tsx, lazy-icon.tsx	byte-identical, so no anchor moved
component.zod.ts metric icon	plugin-dashboard index.tsx, ObjectMetricWidget.tsx, MetricWidget.tsx	byte-identical
component.test.ts / component.zod.ts kanban limit ×3, quickAdd ×2, navigation	plugin-kanban (all), objectql.ts, ElementDataSourceGate, ListView.tsx, ObjectView.tsx, plugin-kanban.mdx	byte-identical. ⚠️ The quickAdd block's lit control said 6 onCardClick; the identical file counts 11 at both pins (the test-side twin already says 11). It was miscounted, not moved, and is corrected; the verdict rests on the 0
component.zod.ts calendar navigation	ObjectCalendar.tsx (changed, afb228418)	re-read: 921-935, 936 and 938-939 unmoved; 1293→1317, 1218→1237 and 1228-1241→1250-1263, text byte-identical
component.zod.ts object-map header, filter, sort; component.test.ts / component.zod.ts map/gantt/tree rows; functional-completeness.ts map row	ObjectMap.tsx (changed, 2252653d0: only warnOnTopLevelStyleUrl's docblock and dev-warning text)	every anchor from :289 on MOVED 16-19 lines with its text byte-identical (e.g. 814→833, 895→914, 377→396, 382→401, 1058→1077); no read point or key set changed
component.zod.ts gantt / tree ×3 / timeline ×2 headers; functional-completeness.ts gantt, timeline rows	plugin-gantt, plugin-tree, plugin-timeline, record-source.ts, ListView.tsx	byte-identical. The tree's ElementDataSourceGate count was re-taken: tree 0, map/gantt/grid/calendar 3 each (plugin-grid/src/index.tsx changed and still reads 3)
api-methods-batch-conformance.test.ts sys_api_key	ObjectGrid.tsx (changed, 804831c2a + 62597c588)	⚠️ the selection block's CONTENT moved: objectui#10218 rewrote the selection arm to "presence enables; an explicit off wins". The span is now 4024-4051 (hash c88443302d40, was 6133933199… over the last three hops). Re-read, the record's claim holds: with neither selection nor selectable authored, multi-select is enabled only by hasBulkActions, and none of the four views declares selection
dataset.zod.ts measure format	date-display.ts (changed, 516583b54), dataset-format.ts	formatDate now parses via toDisplayDate (date-only values keep their calendar day); the cited style handling is unchanged; 198-233→271-306, 152→225
view.zod.ts view-face limit reach (#19228)	9 files	both instruments re-run: the probe returns 0, and the control returns the same 13 lines across 6 files, line for line. All 9 cited files are byte-identical
view.zod.ts:1975/:1978 ObjectMap anchors (content-asserted)	ObjectMap.tsx	re-READ against 2252653d0: it rewrote the docblock and dev warning ABOVE getMapConfig (the remedy now names the declared map: { style }, carried out as mapStyle). getMapConfig is byte-identical, so 385→404 and 390→409; the quoted getMapConfig comment is intact (:361→:380)
view.zod.ts FormField.span docblock	autoLayout.ts (changed, e026e15f9), form.tsx	resolveColSpan 153→154 (an import added above it). form.tsx is byte-identical, so spanLadderFor still emits the ladder

pnpm exec tsx scripts/check-objectui-pin-citations.ts --verify-anchors reads: all 7 content-asserted anchors pass at 62597c588.

The two sites first reported, then resolved under the second ruling (comment 5793509969, option A)

1. packages/spec/src/kernel/functional-completeness.ts, the calendar row: its claim had become FALSE, so it is RESTATED. The row said the "Calendar configuration required" literal "is the text RENDERED at the pin: objectui#10101 landed AFTER the pin". git merge-base --is-ancestor afb228418 62597c588 → 0 (and → 1 against 87af769e9a3e).
   • At the new pin the screen's copy goes through tt('calendar.configRequired', …) (ObjectCalendar.tsx:1207-1208). Its default and the en pack read the same English sentence; zh, de and the other packs render their own words.
   • The if (!calendarConfig) refusal arm is unchanged (the ObjectCalendar.tsx diff on this hop is copy-only).
   • The row now says what renders, names the arm rather than the English sentence as the thing to match, and cites .objectui-sha = 62597c588. Comment only.
2. packages/spec/src/ui/view.zod.ts FormField.span .describe(): its claim still holds, so the citation is RE-POINTED. The describe now names .objectui-sha = 62597c588.
   • Re-read there: plugin-form's WIDE_FIELD_TYPES is unchanged (it shifted one line; repeater still reaches it through field:grid), and form.tsx is byte-identical.
   • Regenerated with pnpm --filter @objectstack/spec gen:schema && gen:docs, which touches only the two span rows in view.mdx.
   • .changeset/19503-pin-bump-describe-corrections.md (@objectstack/spec: patch) states the correction, after PR chore(objectui): bump the console pin to 87af769e9a3e with 27 re-measured citations, and fold in the two-code sdui-parser port (#17645) #19398's 17429-pin-bump-describe-corrections.md.

Gates

node scripts/pm/dispatch-gates.mjs --repo objectstack-ai/objectstack --commands on head 3fd40bedd derives 124 families. The docs families are added by view.mdx. Per-family exit codes are in the os-dev-report comment on #19503. Summary:

• check:objectui-pin-citations exit 0; pnpm --filter @objectstack/spec check:generated exit 0 (all 15 artefacts)
• run as CI runs it and green: check:react-declaration-parity (MANIFEST=$PWD/sdui.manifest.json … --baseline react-declaration-parity.baseline.json --strict, "no new DECLARATION divergence"); the derived spelling omits the manifest input
• check:skill-examples read exit 3 (PREREQUISITE NOT MET: @objectstack/client-react unbuilt) in the sweep; after pnpm --filter '@objectstack/client-react...' build it reads exit 0 (259 prose examples type-check). dispatch-gates --ran: 124 derived, 122 run, 2 NOT-MEASURED, 0 UNRUN
• NOT MEASURED, left to CI: check:dual-build-cjs-loads and check:type-check-debt (both exit 3, PREREQUISITE NOT MET: they need a full workspace pnpm build; this diff changes only comments and migration-rationale strings in spec)
• tests: @objectstack/spec vitest on component.test.ts, api-methods-batch-conformance.test.ts, src/migrations, functional-completeness and src/ui/view*: 22 files, 1562/1562 pass

Changed lines: 560 (+364 / -196, 19 files), under the 5000 threshold. origin/main was merged in twice (d3ff597a before the PR was opened; 16329059b for this round, with view.mdx regenerated from the merged tree as the pre-push os-regen check requires).

Acceptance notes

• component.zod.ts's flat-map guidance block also cites ObjectMap.tsx:382-396 without a pin. That spelling is outside the gate's population, and it now reads 401-415 (byte-identical). It was left as it is: the gate does not name it.
• spec span/colSpan tier prose is anchored to the current .objectui-sha and goes stale at the bump that absorbs objectui#9253 — nothing mechanical holds the ticket #18508 is not absorbed here: its trigger commit bd09957380 is already an ancestor of the OLD pin 87af769e9a3e.

维护者速读(草稿)

• 改了什么:把控制台钉住的 objectui 版本推进到 62597c588(包含权限修复 objectui#10153:record:quick_actions.requiredPermissions 改为失败即拒绝);同时按 PM 裁决 A,重录 sdui-parser 的对齐记录(只动了出处字段),并在新版本上逐条重新核对 spec 里引用 objectui 代码行号的注释。
• 为什么改:当前发布的控制台仍按旧逻辑“失败即放行”,这是 [Decision] requiredPermissions 在 record 块上到底是对象动作还是 ADR-0066 能力？—— 已发布的 record:quick_actions 那一半今天是 fail-open #19186 的解锁前提;钉住版本一动,引用旧版本的注释门禁必然变红,所以这些注释必须随版本重新测量,不能只替换哈希。
• 风险与代价(含回滚):不改任何 schema 或导出,已写的元数据接受/拒绝结果完全不变;唯一对外文字变化是 FormField.span 的 describe 把所引版本改为新钉住版本(说法本身仍成立,附 spec patch changeset)。控制台行为只随 objectui 这 37 个提交变化(27 条带发布说明)。日历那一行说法在新版本上已不成立(提示文案改为按语言翻译),已按第二次裁决改写。回滚:还原本 PR 即可。
• 席位意见:
• 你要做的:本 PR 需要在最终 head 上做一次同档合约审查,记录后才能排队;PR 保持草稿。

Dev session: session_01852ixPjE785Lt8MUeYGDEb (PM seat domain:devx#1, session_01VDtqoecgES7ScQYGbFVDRv).

🤖 Generated with Claude Code

https://claude.ai/code/session_01852ixPjE785Lt8MUeYGDEb

---

Generated by Claude Code

[github-actions]

📓 Docs Drift Check

This PR changes 2 package(s): @objectstack/sdui-parser, @objectstack/spec, touching 13 documentable anchor(s). ⚠️ 1 changed file(s) yielded no anchor (packages/sdui-parser/objectui-lockstep.json), so the pages documenting them are NOT COVERED by this run — this is not a clean bill of health for those files.

2 hand-written doc(s) NAME something this change touched and may need an implementation-accuracy re-verification:

• content/docs/data-modeling/analytics.mdx (via DatasetMeasureSchema (symbol, a top-level const))
• content/docs/protocol/objectui/layout-dsl.mdx (via ComponentPropsMap (symbol, a top-level const object))

⛔ 4 release-owned page(s) also name something this change touched. These are read-only:

• content/docs/releases/v15.mdx (via PageTabsProps (symbol, a top-level const object))
• content/docs/releases/v17/17-1.mdx (via ComponentPropsMap (symbol, a top-level const object))
• content/docs/releases/v17/17-3.mdx (via ComponentPropsMap (symbol, a top-level const object))
• content/docs/releases/v17/17-4.mdx (via ComponentPropsMap (symbol, a top-level const object))

content/docs/releases/ is RELEASE-OWNED (AGENTS.md "Documentation Guardrails"): release
notes are written centrally at release time, and a code PR that edits them is the exact PR
that guardrail exists to stop. They are still audited — read-only. If one of them is actually
wrong, file an issue or open a dedicated docs-only PR; do not edit it here.

What this run could not see

• 1 changed file(s) yielded no anchor (packages/sdui-parser/objectui-lockstep.json) — pages documenting those are invisible to this run
• 2 name(s) were too generic to anchor anything (single lowercase words)
• the SDK route bridge reached 60 of 215 client-bound route-ledger rows — the other 155 have no registrar path: tail to select them, so pages documenting THEIR client methods cannot appear above, on this or any run. Of those 155: 0 are remediable by widening that discovery convention (an in-repo file declares the path; the convention did not scan it); 55 are structural — on a ledger where NOT ONE row is declared in-repo, so no discovery change reaches them at any price; 100 are undecided (no in-repo declaration, on a ledger that has other in-repo registrars — absence and an unreadable spelling are not distinguishable here). The rows themselves: node scripts/docs-audit/affected-docs.mjs --bridge-coverage
• a page that states a rule by its inputs shares no identifier with the emitter that implements the rule, so an emitter-only diff cannot list it — not on this run and not on any run. Measured on fix(driver-sql): emit varchar(maxLength) for a text field a declared index keys on #11430: content/docs/protocol/objectql/types.mdx documents the text-family column mapping by the ObjectQL type names it maps FROM (text / textarea / html) while the diff changed createColumn; it went unlisted, and it was the page that diff falsified, in four places. No shared token exists to detect this on, so a rule your change carries has to be re-read by hand in the pages that restate it.
• a key NAME is not a key, so the hand re-read the line above prescribes can land on the wrong schema. The same spelling is authorable on one governed type and a [REMOVED] tombstone on another for each of active, aria, joins, objects, template, tools and version (censused on [finding] tools is a key on BOTH AgentSchema (tombstoned, dead) and SkillSchema (live, cloud-attested), so a name-based search attributes skill examples to the agent key — it produced a false stop-the-line alarm on PR #19059 #19093 over the liveness ledger's governed types, top-level keys); nothing in a search result distinguishes the two, so a grep hit on a LIVE example reads as evidence about the DEAD key. Measured on fix(spec): the agent.tools liveness row says dead — it claimed live on a key the schema tombstoned #19059: content/docs/ai/agents.mdx was reported as contradicting the agent.tools tombstone over its tools: example at :161, which is inside the defineSkill({ block opened at :155 — the page was already correct. Settle ownership by PARSING the value against both schemas, never by the name: that literal PASSES SkillSchema, and as an AgentSchema it FAILS at tools with the tombstone prescription. ⛔ These names are not the whole class — a key retired through a .strict() guidance map leaves no tombstone in the walked shape and none of them here (tool.category, live as AIToolDefinition.category).

Coarse fallback — 137 page(s) merely mention a changed package (the pre-#9192 predicate, kept for the deliberately-wide backstop): node scripts/docs-audit/affected-docs.mjs --json 8cbc3c0084a3c3c8ef5e59763ab3be96aa1a6f07 → packageMentionDocs.

Which tree this was computed on

This run read content/docs from 1e1ae74246e912409fef2feaad7f3d0215d8743f — the merge of head 2a768071114e39151f9231823700d17c39f3b7f3 into base 8cbc3c0084a3c3c8ef5e59763ab3be96aa1a6f07, which is what actions/checkout gives a pull_request run. Not the PR head.

A worktree cut from an older main holds a different content/docs, so re-deriving there can legitimately return a different list — that is a different tree, not a wrong row. To answer on the same tree:

# while this PR is open — GitHub drops the merge commit once it closes
git fetch origin 1e1ae74246e912409fef2feaad7f3d0215d8743f && git checkout 1e1ae74246e912409fef2feaad7f3d0215d8743f
# afterwards, rebuild it from the two parents, which stay fetchable
git fetch origin 8cbc3c0084a3c3c8ef5e59763ab3be96aa1a6f07 2a768071114e39151f9231823700d17c39f3b7f3 && git checkout -B drift-repro 8cbc3c0084a3c3c8ef5e59763ab3be96aa1a6f07 && git merge --no-ff 2a768071114e39151f9231823700d17c39f3b7f3

node scripts/docs-audit/affected-docs.mjs --json 8cbc3c0084a3c3c8ef5e59763ab3be96aa1a6f07

⚠️ That checkout carried uncommitted changes, so the commit above does not fully identify what was read.

Advisory only, and a precision-first one (#9192): a page is listed because it names a
symbol, wire route or SDK method this diff touched — not because it mentions a changed
package. Each row says which anchor put it there, so a wrong row is reportable rather than
merely annoying. To re-verify, run the docs-accuracy-audit workflow scoped to these files:
node scripts/docs-audit/affected-docs.mjs 8cbc3c0084a3c3c8ef5e59763ab3be96aa1a6f07 → pass the list as
args.docs, on the commit named under Which tree this was computed on.

[objectstack-fleet]

Contract review

Served-tier: CONTRACT_REVIEW_TIER
Head-sha: 3fd40bedd80463437f8435048bdc1c73da14d428

① Derived judgments

• Executable-surface check — PASS. Every touched non-test spec file was comment-stripped (block + line comments, strings preserved) at origin/main and at the head, and the results diffed:
  • packages/spec/src/ui/component.zod.ts (+128/−79): stripped diff EMPTY — every hunk is JSDoc/comment prose (PageTabs, PageAccordion, element:button, metric, kanban, calendar, object-map header, map filter/sort, gantt, tree, timeline, ComponentPropsMap).
  • packages/spec/src/ui/dataset.zod.ts: stripped diff EMPTY.
  • packages/spec/src/kernel/functional-completeness.ts: stripped diff EMPTY (calendar, gantt, timeline, map rows — comment only).
  • packages/spec/src/ui/view.zod.ts: against the PR's merge-base afc3b649 the stripped diff is exactly ONE line — the FormField.span .describe() at head :3264, whose only token change is 87af769e9 → 62597c588. (Against the origin/main tip, extra columns/hiddenFields/fieldOrder describe lines are main drift from 9dcdb775, landed after the merge-base; merge-tree is clean and mergeable_state is clean.) No schema, default, enum, export or other .describe() moved.
  • Migration entries: all six hunks sit inside the reason: string; no id, key, version or transform changed.
  • packages/spec/src/migrations/registry.ts: mechanical regeneration — its changed lines are byte-identical to the entries' changed lines modulo indent; nothing else moved against the merge-base.
• Test files — no assertion weakened. component.test.ts and api-methods-batch-conformance.test.ts: comment-stripped diff EMPTY for both; the edits are // citation prose only.
• Citation spot-checks at objectui 62597c588072 (full clone; 30+ anchors read, all as claimed), including: the two view.zod.ts ObjectMap anchors (ObjectMap.tsx:404 ObjectMapConfigSchema.safeParse(config), :409 the style fallback — byte-identical to :385/:390 at 87af); the calendar row (ObjectCalendar.tsx:1200 if (!calendarConfig) → :1206-1208 tt('calendar.configRequired', …); en.ts:711 carries the English sentence, zh.ts:598 / de.ts:579 their own; afb228418 is an ancestor of the new pin and not of the old); the calendar navigation read points; FormField.span (autoLayout.ts:154 resolveColSpan, WIDE_FIELD_TYPES identical at both pins, form.tsx byte-identical); the grid selection block hashes in api-methods-batch-conformance.test.ts; dataset.zod.ts measure format; the byte-identity roster; corpus counts re-taken at both pins (dark tokens 0 at both; onCardClick = 11 at both — the "6" correction is right).
• packages/sdui-parser/objectui-lockstep.json: only rev, revDate and recordedAgainstPin changed; revDate equals the new pin's commit date. Grammar untouched.
• The pin: 28b065800c7d (the fail-closed fix) is an ancestor of 62597c588072 (exit 0, non-shallow clone); 87af769e9a3e is an ancestor of the fix; 33 commits fix → pin. .objectui-sha and scripts/sdui-manifest.record.json carry the full sha.
• No stale asserting citation: zero matches of the gate's asserting pattern for the old pin remain in packages/spec/src; the one leftover 87af769e9a in a .pin.test.ts is the historical spelling.
• Clause-②: no is right: no accept-set change, no exported name moved, no public surface widened.

② Semver level

• .changeset/console-62597c588072.md → @objectstack/console: minor: correct and truthful (30 objectui changesets in range, 0 major; lists the fail-closed fix).
• .changeset/19503-pin-bump-describe-corrections.md → @objectstack/spec: patch: correct — one describe sentence re-points a pin citation. No major anywhere.

③ Boundary flags

• No governed surface and no release-owned tree in the 19-file list.
• content/docs/references/ui/view.mdx: exactly the two span rows, generator output only.
• origin/main has moved since the merge-base; merge-tree is clean; a later re-merge will need gen:docs from the merged tree, which the pre-push os-regen hook enforces.
• Check-runs on 3fd40bedd: 42 total, 39 success, 3 skipped, 0 failures.

Implemented-by: session_01852ixPjE785Lt8MUeYGDEb
Reviewed-by: session_01VDtqoecgES7ScQYGbFVDRv

VERDICT: PASS

Isolated at-tier reviewer, adopted by the domain:devx#1 seat.

---

Generated by Claude Code

[objectstack-fleet]

Regen-provenance: 5794397942 · 3fd40bedd80463437f8435048bdc1c73da14d428 → 2a768071114e39151f9231823700d17c39f3b7f3 · git diff --name-only 3fd40bedd..2a7680711, minus paths whose blob equals origin/main → content/docs/references/ui/view.mdx (regenerated), packages/spec/src/migrations/registry.ts, packages/spec/src/ui/view.zod.ts (main's delta, verbatim)

The hop was produced by bash scripts/pm/os-regen-merge.sh (merge commit 557f561aa, parents 3fd40bedd + origin/main 8cbc3c008). The regeneration commit 2a7680711 changes one file, content/docs/references/ui/view.mdx, via gen:schema + gen:docs. The os-regen driver had kept this branch's side and dropped main's fieldOrder / hiddenFields describe rows; the regenerated file equals origin/main's except for this PR's two FormField.span rows.

Of the 52 paths in git diff --name-only 3fd40bedd..2a7680711, 49 are byte-identical to origin/main. The other two, registry.ts and view.zod.ts, were text-merged by git without conflict. For each, the hop's +/- lines are identical to main's own delta (git diff afc3b649 origin/main compared with git diff 3fd40bedd 2a7680711, matching sha1). No other change is in the push.

Step 4 on 2a7680711 (spec build VERDICT exit 0, then gen:docs and gen:migration-registry, which produced no further diff) — each exit 0:

• check:generated
• check:objectui-pin-citations
• check:docs
• check:api-surface
• check:authorable-surface
• check:sdui-lockstep
• check-sdui-manifest
• check:future-spec-major
• check:objectui-changeset

Main's sibling migration entries are present at the same counts in registry.ts.

Dev session session_01852ixPjE785Lt8MUeYGDEb.

---

Generated by Claude Code