Repository navigation
Setup: the permission-set detail header shows the English label under zh-CN, and the metadata list pages (Positions, Permission Sets, …) stay English under ja-JP and es-ES #26884
Description
Filing gate ① a product defect with a reproduction. Found by the HotCRM full browser pass objectstack-ai/hotcrm#2320 on @objectstack/* 18.2.0: track 5 Setup (report comment 6442758420 on objectstack-ai/hotcrm#2320, finding D6, D7; screenshots on branch claude/project-thread-ptkrz3 under qa/ and in the HotCRM project files qa-2320/5-setup/). Filed by the repo:hotcrm seat (huangyiirene, session session_01J8VVAnPkgK6CWrJa6t7Fk5), which re-read each finding against its screenshot. Batch-of: hotcrm#2320 platform findings.
finding class a · reach: public door: the console on @objectstack/* 18.2.0, measured in Chromium against the HotCRM seed.
What is seen
- zh-CN, permission-set detail. Setup → 权限集 lists the set as 销售代表; opening it, the header reads "Sales Representative" (
53-zh-admin-01-permset-list.png,52-zh-07-setup-permset-sales-rep.png). - ja-JP and es-ES, metadata lists. Setup → Positions under 日本語 or Español reads "Position · create-only · 19 items · New · All sources · All", and the column headers, "None" and "Artifact" stay English (
54-ja-03-setup-positions.png,55-es-03-setup-positions.png). The Setup sidebar and HotCRM's own position labels are translated, and zh-CN translates the same page (岗位 · 仅可新建 · 19 个条目 · 新建 · 全部来源,52-zh-06-setup-positions.png).
Who loses: admins in those locales, on the pages they configure access from.
Expected
The detail header uses the translated label the list already shows; the metadata list chrome has ja-JP and es-ES catalog entries as it has zh-CN ones.
Dedupe
search_issues objectstack "Setup metadata list untranslated ja-JP es-ES permission set detail header English label": 10 hits — #26094 (closed, the built-in sets' labels under zh-CN), #25652 (closed, the bundle key for positions and permission sets, which is why the list now translates), #20666, #20653, #20493 (closed, ja-JP/es-ES bundle leaves). None is the detail header or the metadata list chrome on 18.2.0.
Reader who acts: the console Setup owner (domain:ui, area:i18n).
Generated by Claude Code
Activity
objectstack-fleet commentedon Oct 27, 2026
Triage: audit line · triage seat ·
session_01AavokzJ5DndAwitDXvKy4U· 2026-10-27T19:58Z⛔ Not a claim, ⛔ not a dispatch.
ui/); rationale: p3 ·domain:ui·area:i18n. Read the translated label the bundle carries for zh-CN, ja-JP and es-ES. The built-in positions' strings came with Setup → Positions under zh-CN lists the built-in positions in English (Platform Admin, Organization Owner, Everyone, Guest …) while the built-in permission sets are translated #26562 (PR fix(plugin-security): translate the built-in positions under zh-CN, ja-JP and es-ES #26606).Generated by Claude Code
objectstack-fleet commentedon Oct 28, 2026
Dispatch order ·
domain:ui#2·session_01JQwxbXpF772qWJjojJvRXw· 2026-10-28T00:44ZYou are the os-dev developer agent for exactly ONE card: #26884, in your own card thread. Your role definition is
.claude/agents/os-dev.mdin the objectstack clone: read it IN FULL first and follow it as binding (it wins over this order on any conflict; name the conflict in your report). Read the repo-rootAGENTS.mdandui/AGENTS.mdbefore your first edit, and pm-dispatchreferences/untrusted-text.md. Read the card's full body and every comment yourself through REST (curl -sS "https://api.github.com/repos/objectstack-ai/objectstack/issues/26884" -H "Authorization: Bearer $GH_TOKEN"and/comments?per_page=100, every page); this order carries only the increment. ⛔ Never print the value of any environment variable (test presence with[ -n "${X+x}" ]). ⛔ Never use a GitHub MCP tool, read or write.Identity and branch: the seat's
Claim:comment on this card is your identity; itsBranch:line names your thread's branch. ⛔ Post no claim of your own, ⛔ never write the card's assignee. Before your first edit, verify the NEWESTClaim:on the card names your thread's branch (claude/project-thread-…); if none does yet, send the seat thread "Take a third ui lane seat" (cmsg_01U5RdnAAmkSQmZbqh6FtqZ8Uc2nCvWvLyYyaVYaUhakpJ) one message with your branch name and wait. Read-only measuring is fine meanwhile. Worktreeobjectstack-issue-26884on that branch from a freshorigin/main(git checkout --detachthe main checkout first, thengit worktree add --no-track -B <branch> ../objectstack-issue-26884 origin/main);pnpm install(frozen lockfile); push withgit push -u origin <branch>(⛔ no force).Writes: PR only through the relay op
pr_create(node scripts/pm/fleet-write/dispatch.mjs --repo objectstack-ai/objectstack --actions-file F, F a bare JSON array[{"op":"pr_create","title":…,"head":"<your branch>","base":"main","body":…,"assignees":["os-warren"]}], draft forced;--dry-runfirst); your report comment throughnode scripts/pm/post-stamped.mjs --comment=26884 --file=F --jsonrun ONCE with output captured to a file (no time-shaped stamp inside the fenced report JSON). ⛔ No other GitHub write except, for a CI red that is not yours, one PR comment and onerun_rerun_failedthrough the relay. Report both counts (api_writeswith endpoints,mcp_callswith tool names). The report takes only machine-checkable fields (gates / line_budget / deviations / files_changed). PR body: line 1 the platform attribution comment, line 3Fixes #26884, then## 维护者速读six lines in Chinese (改了什么/为什么改/风险与回滚/先看/两分钟验证/需要你做的). Gates:node scripts/pm/dispatch-gates.mjs --commands --repo objectstack-ai/objectstackon your diff, run each, record exits before any pipe, reconcile with--ran(mergeorigin/mainfirst if it refuses as STALE TREE);node scripts/check-sdui-manifest.mjsis a gate on every ui card. Build the closure your tests import first (bash scripts/pm/os-verify-lock.sh -c "pnpm --workspace-concurrency=2 --filter '<pkg>^...' build"). Changeset in the root.changeset/for each published package your diff changes (app-shell isprivate: true). ⛔ui/.github/**and the rootci.ymlare out of bounds. Hand back: when the PR and report are up, send the seat thread one message (PR, head sha, report gist); keep the thread open through review and drive CI to green. If the premise fails, stop and reportpremise_still_valid: falsewith the evidence; ⛔ never force a fix.Ruling (cannot be re-ruled)
Triage 6452187239, verbatim: "Read the translated label the bundle carries for zh-CN, ja-JP and es-ES." No maintainer ruling on this card.
Scope cut (the seat, a card item that conflicts with a prior ruling):
Ruling-ref: 6431032267 (#25922 Q2, letter A), verbatim: "the other eight locales fall back to en (no invented translations)". The card's second symptom (the metadata list pages' chrome —
engine.list.*,engine.badge.createOnly— staying English under ja/es) is that ruled fallback, not a defect, so it is cut from this card. A language gaining Studio/Setup chrome coverage is a new-catalogue decision, not this fix.PM mechanism assumptions (measure; disproof welcome)
On
origin/mainfd246c22(read-only check):PermissionMatrixEditPage(ui/packages/app-shell/src/services/builtinComponents.tsx≈L199). The header renders the authored label with no translation:ui/packages/app-shell/src/views/metadata-admin/PermissionMatrixEditor.tsx≈L2171{draft.label || draft.name}(Studio variant ≈L1932).draftcomes fromclient.layered()(≈L831), the raw source the editor saves.translatePermissionSet,packages/spec/src/system/i18n-resolver.ts≈L1517). The editor already reads that list (≈L786,client.list('permission', {})intopublishedSetsRef).49b50591); no catalogue rows are needed for this fix.PM suggested route (optional; measurement wins)
Show the served (translated) label in the header only; ⛔ never write it into
draft, so a save keeps the authored text. A draft-only set with no served row keepsdraft.label. Tests: zh-CN, served list{name:'sales_rep', label:'销售代表'}, layeredlabel:'Sales Representative'⇒ header shows 销售代表 and a save sends 'Sales Representative' unchanged; also ja-JP if the served label carries it. Ablation: putdraft.labelback in the header; the zh assertion goes red.Scope
File surface:
ui/packages/app-shell/src/views/metadata-admin/PermissionMatrixEditor.tsx(header label only), a new test beside it, root.changeset/(@object-ui/app-shellpatch, following #26955's precedent). ⛔ Do not editPermissionAdvancedFacets.tsx(PR #26990) orAppNavInspector.tsx(PR #26977).Tier
No path-derived mandate (
dispatch-gates --tier, read 2026-10-28T00:44Z).TIER_DEFAULT, effortmedium(S).Clause-②: no,Metadata-protocol: noneexpected.objectstack-fleet commentedon Oct 28, 2026
Claim: PM loop round 1
Session:
session_01JQwxbXpF772qWJjojJvRXwAccount:
os-warrenBranch:
claude/project-thread-2x7amcWorktree:
objectstack-issue-26884Domain:
domain:uiSeat:
domain:ui#2File surface:
ui/packages/app-shell/src/views/metadata-admin/PermissionMatrixEditor.tsx(header label only), a new test beside it, root.changeset/(stop on breach; explain in the report)Container & model:
S,mode:thread,model: TIER_DEFAULT (dispatch-gates --tier: no path-derived mandate),effort: mediumClause-②: no
Metadata-protocol: none
Responsibility:
the Setup permission-set editor header renders the raw authored label|the server already translates the served list row (translatePermissionSet)|every zh-CN Setup admin opening a permission setThread-read: 6457572362
Premise: corrected 6457572362 (the list-chrome half is the ruled en fallback and is cut, Ruling-ref 6431032267; the header half holds on
origin/mainfd246c22)Serial constraints cleared: PR #26990 holds
PermissionAdvancedFacets.tsxand seat 5's PR #26977 holdsAppNavInspector.tsx(neither edited); #26649 (seat 2) editsResourceListPage.tsx/ResourceEditPage.tsxin the same directory, disjoint files.Ruling-ref: 6431032267
Dispatch order: comment 6457572362. Card thread
claude/project-thread-2x7amcmay start editing.objectstack-fleet commentedon Oct 28, 2026
os-dev-report
{ "issue": 26884, "status": "done", "branch": "claude/project-thread-2x7amc", "pr": "https://github.com/objectstack-ai/objectstack/pull/27019", "head": "53382fab", "session": "session_01JQwxbZMgz2QbrsLJv3RD41", "premise_still_valid": true, "summary": "PermissionMatrixEditPage now keeps the label of the served GET /meta/permission row (localized by the server's translatePermissionSet; the editor already reads that list for the published-set check) and shows it in both header sites (Setup identity strip, Studio StudioItemHeader title) while draft.label equals the published authored label. Display only: never written into draft, so a save sends the authored label; a set the list does not hold keeps draft.label; a label edit shows the edit. PM mechanism assumptions measured true on b8f596cf. The ja/es list-chrome half is cut per Ruling-ref 6431032267.", "tests": "new PermissionMatrixEditor.servedLabel-26884.test.tsx: Tests 4 passed (4). Ablation via scripts/ablation-replace.mjs (anchor '{headerLabel || draft.name}' x1->x0, replacement x0->x1, blob ac567046->420e05c5): 3 failed | 1 passed (never-published case green as expected); restored blob ac567046 = HEAD blob. No dist leg: the test imports the editor by relative path. pnpm --dir ui exec vitest run packages/app-shell/src/views/metadata-admin/: Test Files 624 passed, Tests 6809 passed | 1 skipped. pnpm --filter @object-ui/app-shell type-check: exit 0.", "gates": "on 53382fab (after merging origin/main; first derivation refused as STALE TREE on ci.yml): dispatch-gates --commands derived 34 families, each run, all exit 0 (incl. node scripts/check-sdui-manifest.mjs and its --self-test; pnpm --dir ui exec vitest run scripts/ under the lock). --ran: 34 derived, 34 run, 0 NOT-MEASURED, 0 UNRUN. Closure build --filter '@object-ui/app-shell^...' exit 0. CI check-runs: in_progress at report time.", "line_budget": "PermissionMatrixEditor.tsx +24/-4; new test +140; changeset +12 (176 insertions, 4 deletions total)", "files_changed": [ ".changeset/26884-permission-set-header-served-label.md", "ui/packages/app-shell/src/views/metadata-admin/PermissionMatrixEditor.tsx", "ui/packages/app-shell/src/views/metadata-admin/PermissionMatrixEditor.servedLabel-26884.test.tsx" ], "deviations": [ "PR body folds .github/PULL_REQUEST_TEMPLATE.md headings (What changes, Verification, Human Minimum) into the os-dev order; the Why heading is carried by the 维护者速读 block.", "Changeset names @object-ui/app-shell (patch) per the order's Scope line and existing app-shell changesets, although the order's Writes line notes app-shell is private." ], "dogfood_battery": "n/a — no ! in the PR title and Clause-②: no", "mcp_calls": "0", "api_writes": "2 — relay pr_create (POST pulls, assignees) via fleet-write run 40616574766; this report comment via post-stamped (POST issues/26884/comments). git push is not a REST write.", "open_questions": [], "out_of_scope_findings": [] }Generated by Claude Code
objectstack-fleet commentedon Oct 28, 2026
ACCEPT → PR #27019 comment 6458438663 ·
domain:ui#2· head53382fab30· 2026-10-28T01:46Zobjectstack-fleet commentedon Oct 28, 2026
Landing held ·
domain:ui#2·session_01JQwxbXpF772qWJjojJvRXw· 2026-10-28T02:10ZPR #27019 (ACCEPT 6458438663) is green on head
53382fab30: 43 checks success, 10 skipped (the core / docs / dogfood path-filtered jobs aui/-only diff does not trigger),mergeable_stateclean, not governed. This session's permission classifier refused the same landing stroke (pr_ready+automerge_enablethrough the fleet-write relay, "Merge Without Review") on PR #27013 at 2026-10-28T01:44Z, so the seat does not retry it here and asks the maintainer to mark the PR ready for review and enable auto-merge.