Skip to content

fix(app-sdk): show "New session" in an untitled ChatEmbed header - #19887

Merged
buluoray merged 2 commits into
mainfrom
fix/embed-untitled-header
Oct 16, 2026
Merged

buluoray merged 2 commits into
mainfrom
fix/embed-untitled-header

Conversation

@buluoray

@buluoray buluoray commented Oct 13, 2026 •

Copy link
Copy Markdown
Contributor

Problem / Motivation

Goal: An embedded chat that has no title yet shows "New session" in its header instead of an internal id.

ChatEmbed's header rendered {title || slotKey}, so until a session got a title the header showed the raw slot key, for example coder-abc123. That is an internal identifier that means nothing to the user.

Why it matters

Every app that mounts ChatEmbed on a fresh session shows this id as the first thing the user reads, in every language.

Not a goal

  • Generating or fetching a better title for the session.
  • Any other ChatEmbed behaviour (send, stop, tool details, queue, attachments, navigation).
  • Changing the frameless mode, which has no header at all.

What changed (motivation → approach → change)

  • Symptom: untitled header shows coder-abc123.
  • Cause: the header fell back to slotKey when title was empty.
  • Change: the fallback is now the catalog string appSdk.chatEmbed.untitled_session_title ("New session"). The slot key moves to the header's title attribute, so it is still visible on hover for debugging.
  • A new key was added rather than reusing components.sessionGridView.new_session, because that one is a button label (an action) and website/docs/i18n-catalog.md asks for one key per grammatical role. The translations use each locale's existing word for "session"; en-XA was regenerated with npm run i18n:pseudo.
  • docs/app-kit/api-reference.md now documents the header's untitled behaviour.
  • docs/system-specs/modules/app-kit-platform.md now pins the same fallback for fresh and resumed untitled chats, later title updates on the mounted header, the hover title, and frameless behavior.

Backwards compatibility

Compatible: no prop or API changes; only the visible text of an untitled header changes, and the slot key is still exposed as the header's title attribute.

Removes nothing: the slot key is no longer header text but remains on the same element as its hover title; no prop, key or route is removed.

Tests

Review follow-up at 1d8d35119e7ff493a7846ce092295975239cb25b:

  • Added two ChatEmbed.test.tsx polling cases, for fresh and resumed untitled sessions. Each checks that a later title replaces the fallback on the same mounted header.
  • Restoring exactly {title || slotKey} makes both new cases fail. All 58 related tests pass across 5 files, including shuffled order (seed 19887); TypeScript, touched-test ESLint, and 53 flake-pattern ratchet tests pass. Full suites were not run locally for this follow-up.

Initial implementation validation:

  • ChatEmbed.frameless.test.tsx — new case: an untitled slot shows "New session", the header's title attribute is the slot key, and the slot key is not visible text. The frameless case now asserts the header element is absent.
  • ChatEmbed.test.tsx — the old "shows slotKey as fallback" case now asserts the new fallback.
  • Mutation check: restoring {title || slotKey} turns both of those tests red.
  • Ran: npx tsc --noEmit (pass), eslint on the touched files (pass), npm run lint:i18n (pass), npm run i18n:check (pass), vitest run src/test/ChatEmbed (5 files, 56 tests pass), vitest run src/i18n (47 files, 769 tests pass).

Manual verification

Rendered the real ChatEmbed in an isolated Vite page with fake slot data (empty title), before (main) and after, in dark, light, a 320px-wide box, and zh-CN. The header reads "New session" / "新会话" after the change and coder-abc123 before; nothing overlaps or truncates at narrow width.

Screenshots / video

Before (main), dark:

before dark

After, dark:

after dark

More variants (after: light, narrow, zh-CN)

after light

after narrow dark

after zh-CN dark

Related Issues

Deferred siblings, one issue per fallback site. All nine additional sites were verified on main at 567df9af4d6f86a170783b1b64c2e0a0f636572a; none was skipped. These are follow-ups, not fixes included in this PR.

Review site Current main site Tracking issue
ChatPane.tsx:535 website/src/components/ChatPane.tsx:219 #20871
SessionGridView.tsx:356 website/src/components/SessionGridView.tsx:356 #20873
SessionsPage.tsx:299 website/src/pages/SessionsPage.tsx:299 #20875
ChatSidebar.tsx:5212 website/src/pages/chat-sidebar/HistoryPane.tsx:286 #20876
ChatPage.tsx:6459 website/src/pages/ChatPage.tsx:2245 #20877
MembersPage.tsx:5286 website/src/pages/members/MembersPage.tsx:5283 #20878
sessionsProvider.ts:174 website/src/components/commandPalette/providers/sessionsProvider.ts:174 #20879
recentsProvider.ts:304 website/src/components/commandPalette/providers/recentsProvider.ts:304 #20880
recentsProvider.ts:392 website/src/components/commandPalette/providers/recentsProvider.ts:392 #20881

Additional sites verified on main 0694e63a87ffb40a332ff3664b803ba71ea4bed1 after the latest First Principles review:

Main site Tracking issue
website/src/components/SessionRefStrip.tsx:58 — attached-reference chip #20907
website/src/pages/members/MembersPageProfile.tsx:560 — driving-session row #20908
website/src/apps/command-bar/CommandBarOverlay.tsx:1044 — attention row #20909
website/src/apps/command-bar/CommandBarOverlay.tsx:1085 — recent row #20910

These are source-verified, not browser-reproduced. The attention row requires a pill status; recent rows exclude empty new slots and attention slots. Their issues preserve those filters. No sibling code changes are included here.

Existing follow-ups: #20348 (remote crew header), #20349 (session tab label), #20362 (ChatEmbed loading and older-untitled wording).

The sidebar history-row implementation moved from ChatSidebar.tsx into chat-sidebar/HistoryPane.tsx; the same key fallback remains. ChatPane, ChatPage, and MembersPage also have updated line numbers. The recents-provider issues distinguish blank current titles from older blank titles with nonempty previews; they do not claim filtered-out history entries are displayed.

Pattern harvest

Loading and age-aware titles are tracked separately in #20362 (tier:T1, unassigned). This PR still shows "New session" before the first title fetch and on resumed untitled chats with prior messages; it does not claim to solve either wording limit.

Rule candidate: review-prompt
Pattern: a session's internal key used as the display-text fallback for a missing title (title || key).

At rebased head dccbfece803fcb2d6c57af651332e1a4b88ca57f, the review's broader inventory contains 11 remaining fallback sites across 10 files, beyond the ChatEmbed header fixed here:

  • website/src/pages/chat/crew-window/CrewChatWindow.tsx:300 — remote crew header; tracked in fix: hide untitled slot ids in the remote crew chat header #20348.
  • website/src/components/SessionTabStrip.tsx:232 — session tab label; tracked in fix: hide untitled slot ids in session tab labels #20349.
  • website/src/pages/SessionsPage.tsx:299 — session row label.
  • website/src/pages/ChatSidebar.tsx:5212 — history row label.
  • website/src/components/SessionGridView.tsx:356 — grid label.
  • website/src/pages/ChatPage.tsx:6459 — history row label.
  • website/src/components/ChatPane.tsx:535 — pane title fallback.
  • website/src/pages/members/MembersPage.tsx:5286 — driving-session row label.
  • website/src/components/commandPalette/providers/sessionsProvider.ts:174 — search-result title fallback.
  • website/src/components/commandPalette/providers/recentsProvider.ts:304,392 — current and older result-title fallbacks. Rendering has separate empty-slot/history filtering; these fields are not proof that every empty session exposes its key.

All 11 sites are deferred, not fixed here. #20348 and #20349 track the two originally named siblings; the nine remaining sites now each have their own issue in Related Issues above, with locations reverified on current main. A shared fallback policy needs separate design and regression coverage. This PR fixes only ChatEmbed and claims no shared solution.

The 11-site list above records the earlier review, not an exhaustive repository inventory. The latest review adds four source-verified sites across three files (#20907–#20910), linked in Related Issues. Those follow-ups remain outside this PR.

Checklist

  • At most two commits (one is the norm), with a Conventional Commits title (feat|fix|docs|style|refactor|perf|test|chore|ci|build|revert: ...)
  • Existing tests pass and new tests added for new functionality
  • Self-review completed; code follows project style guidelines
  • Documentation updated (if applicable)
  • No secrets, credentials, or internal references in the diff

Contribution License Agreement

Contributions are accepted under the project's Apache-2.0 license; see the Licensing section of CONTRIBUTING.md.

@buluoray
buluoray requested review from a team and KiroCrewCaptain as code owners October 13, 2026 19:01
@buluoray
buluoray requested a review from mbriones98 October 13, 2026 19:01
@github-actions github-actions Bot added the readiness: checking Automated validation is still running label Oct 13, 2026
@github-actions

github-actions Bot commented Oct 13, 2026 •

Copy link
Copy Markdown
Contributor

Design Review (Opus 5.5) — ✅ PASS

Design-level review of b266889d88973b8644ff2e081d4b55c84670d27c — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS/CONCERNS are advisory.

Design-Verdict: PASS

Check that public/vendor/kirocrew-app-sdk.mjs gets rebuilt from this source, so apps that load the vendored SDK also show "New session" instead of the slot key.

[DESIGN-REVIEWED] b266889

Description read

[DESCRIPTION-READ] 976c63fb1a8196ef1505bcd1a16294275360f98b5338a942dfe67a309065de84

That sha256 names everything this verdict read: the PR description, plus the 1 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with .github/scripts/pr-description-capture.sh over the same inputs; if it differs, the description or that evidence changed after this verdict was formed and any finding drawn from either is unproven.

@github-actions

github-actions Bot commented Oct 13, 2026 •

Copy link
Copy Markdown
Contributor

First Principles Review (Opus 5.5) — ✅ PASS

Premise-level review of b266889d88973b8644ff2e081d4b55c84670d27c — why this exists and whether the shipped surface is the smallest honest version. Updated in place on each push. A BLOCK verdict blocks PR readiness; PASS/CONCERNS are advisory.

First-Principles-Verdict: PASS

Before merging, check that "New session" is acceptable wording on resumed untitled chats that already have messages. #20362 is deferring that wording problem.

What this change ships

Inventory (6 items) — 6 justified

Intent: someone using an app that embeds a chat wants a readable header on an untitled session, not an internal id like coder-abc123. This is a FIX.

  1. An untitled framed ChatEmbed header now reads "New session" instead of the slot key. — justified
  2. The slot key moves from header text to the header's hover tooltip. — justified
  3. A new catalog key, appSdk.chatEmbed.untitled_session_title, is translated in 13 locales. — justified
  4. A chat-embed-title test hook is added to the header span. — justified
  5. The api-reference doc and the app-kit-platform spec now describe the untitled header. — justified
  6. The pinned test "shows slotKey as fallback when no title" is rewritten to assert the new fallback. — justified

[FIRST-PRINCIPLES-REVIEWED] b266889

Description read

[DESCRIPTION-READ] ea776632381d040765d3a795a464bda8e1cf7799c7902d831a3f04de71e49f17

That sha256 names the PR description this verdict read. Recompute it with .github/scripts/pr-description-capture.sh; if it differs, the description changed after this verdict was formed and any finding drawn from the description is unproven.

@github-actions

github-actions Bot commented Oct 13, 2026 •

Copy link
Copy Markdown
Contributor

Opus 5.5 Review — ✅ no blocking findings

Reviewed b266889d88973b8644ff2e081d4b55c84670d27c — this comment is updated in place on each push.

Review details

No findings.

[OPUS-REVIEWED] b266889

Verdict parsed from the review's SHA-scoped output markers for commit b266889d88973b8644ff2e081d4b55c84670d27c.

False positive or not applicable? A repository writer can comment:
/ai-review override fable b266889d88973b8644ff2e081d4b55c84670d27c: <one-sentence reason>

@github-actions

github-actions Bot commented Oct 13, 2026 •

Copy link
Copy Markdown
Contributor

UX Review (Opus 5.5) — ✅ PASS

UX-level review of b266889d88973b8644ff2e081d4b55c84670d27c — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS/CONCERNS are advisory.

UX-Verdict: PASS

Check one case before merge: a resumed untitled chat that already has messages will still be labelled "New session" (tracked in #20362).

[UX-REVIEWED] b266889

Description read

[DESCRIPTION-READ] 57d88343d8fdd0fc3806526ba0bfe3eefbe03f9cf8c501fb12ba22cac676828e

That sha256 names everything this verdict read: the PR description, plus the 8 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with .github/scripts/pr-description-capture.sh over the same inputs; if it differs, the description or that evidence changed after this verdict was formed and any finding drawn from either is unproven.

@github-actions

github-actions Bot commented Oct 13, 2026 •

Copy link
Copy Markdown
Contributor

GPT 6.1 Review — ✅ no blocking findings

GPT 6.1 completed its review of b266889d88973b8644ff2e081d4b55c84670d27c and found no blocking issues.

This comment is updated in place on each push.

Review details

No findings.
[GPT-REVIEWED] b266889

False positive or not applicable? A repository writer can comment:
/ai-review override gpt b266889d88973b8644ff2e081d4b55c84670d27c: <one-sentence reason>

@buluoray

Copy link
Copy Markdown
Contributor Author

🤖 Kiro Crew Auto-Pipeline [operator: buluoray#19887]: dispositions for the advisory lanes on 7f0e6ef607.

First Principles: symptom-level, 2 siblings. Agreed and verified. CrewChatWindow.tsx builds its header title as slotQ.data?.title || detailQ.data?.title || key, and SessionTabStrip.tsx labels tabs with slot?.title || key. Both are dashboard surfaces outside this PR's Goal (the App Kit embed), so they stay deferred. The Pattern harvest now names both and the pattern instead of saying "not generalizable". No code change.

UX: slot key in the hover tooltip. Kept on purpose. The audit item asked for the slot key to move out of the visible text and stay on the title attribute, so an app author can still find which slot an embed is bound to. The tooltip only appears on a deliberate hover and never replaces the label. Moving it to a data- attribute would hide it from the people who need it. I am not pushing a change for this advisory point. The owner can ask for the tooltip to show only when the title is empty.

UX: "New session" over "Session ready." Both lines are existing copy describing the same chat. Changing the empty-state line is outside this PR's Goal. No change.

@github-actions github-actions Bot added readiness: passed Eligible automated validation passed for the current revision and removed readiness: checking Automated validation is still running labels Oct 13, 2026
CrysisDeu
CrysisDeu previously approved these changes Oct 13, 2026

@CrysisDeu CrysisDeu left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed head 7f0e6ef. Key in all 12 catalogs + en-XA, tests red without the fix, no scope creep. CI green. Approving.

Nits: no test drives one mount from untitled to titled (first resolve {}, then {title:'X'}, advance past the 5000 ms poll); title={slotKey} at ChatEmbed.tsx:329 is unconditional, so titled headers also show the internal id on hover, which the body itself calls meaningless to the user, suggest title={title ? undefined : slotKey}; hi/ru use "सेशन" / "сеанс" in the new string while the status line in the same block uses "सत्र" / "Сессия".

Posted by code-reviewer (Kiro Crew member) on behalf of @CrysisDeu. Full report on file.

@iamwhatever iamwhatever left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This PR changes the ChatEmbed header fallback from {title || slotKey} to a new appSdk.chatEmbed.untitled_session_title ("New session"), keeps the slot key as the hover title, and adds the key to all 12 catalogs plus en-XA. That fixes the root cause for this component. I read the full diff and the tests. Both changed tests assert the new text, assert the slot key is not visible text, and would fail with the old fallback. Checks are green, base is main and mergeable is clean, so the code itself is fine. What blocks approval is deferred work with nothing tracking it. The Pattern harvest names website/src/pages/chat/crew-window/CrewChatWindow.tsx:300 (header falls back to key) and website/src/components/SessionTabStrip.tsx:232 (tab label falls back to key) as deferred, and the author's reply to First Principles agrees. Please open one GitHub issue for each sibling and link both in the description. Also, the replies to the First Principles and UX CONCERNS are one prose comment with no <!-- ai-review-disposition target=first-principles|ux head=7f0e6ef... --> markers, so please repost each item as a marked disposition. Optional, not blocking: title={slotKey} at ChatEmbed.tsx:329 also shows the internal id on titled headers; title={title ? undefined : slotKey} would fit the PR's own goal. Note that #19879 inserts text at the same spot in docs/app-kit/api-reference.md, so whichever PR lands second needs a rebase. — crew-pr-reviewer (deep re-check)

@buluoray

Copy link
Copy Markdown
Contributor Author

🤖 Kiro Crew Auto-Pipeline [operator: buluoray#19887]:

@iamwhatever @CrysisDeu — review follow-up at 1d8d35119e7ff493a7846ce092295975239cb25b.

  • Deferred sibling tracking / First Principles: verified both fallbacks on current main. Opened fix: hide untitled slot ids in the remote crew chat header #20348 for the remote crew chat header and fix: hide untitled slot ids in session tab labels #20349 for session tab labels, each tier:T1 and unassigned. Both are linked in Pattern harvest; the Goal, Why it matters, and Not a goal remain unchanged.
  • Missing title transition coverage: added two cases in website/src/test/ChatEmbed.test.tsx: replaces the untitled label after polling 'a fresh session' without remounting and the same case for 'a resumed session'. Each starts untitled, checks the localized fallback and hidden visible id, drives the real query scheduler to its next poll, and checks the same header node now displays the returned title.
  • Mutation evidence: restoring exactly {title || slotKey} made BOTH new cases fail: expected New session, received coder-untitled. Restored production code before the final run. All 58 related tests pass across 5 files, also with shuffled order (seed 19887). TypeScript, the touched test ESLint check, and all 53 flake-pattern ratchet tests pass. Full suites were not run locally.
  • Design PASS question about resumed untitled chats: the resumed-case test pins the behavior. New session remains the chosen fallback even with earlier turns; it is not an age indicator. The owning App Kit spec now states that explicitly.
  • Optional hover suggestion / UX: retained title={slotKey} as already documented. It remains a deliberate debugging affordance on both titled and untitled headers. Conditional hover is optional in the human review; this test-only follow-up does not change that user-facing choice.
  • UX repeated session wording: the header names the chat, while the existing empty-state line explains readiness and how to start. Changing that status text remains outside this narrowly scoped fallback fix.
  • Hindi/Russian wording nit: the new label uses the catalog wording already used for New session elsewhere; the pre-existing status line uses another synonym. No catalog wording was changed in this follow-up.
  • GPT and Opus: both previous-head reviews reported no findings. New-head reviews are pending.
  • Description overlap with docs(app-kit): state ChatEmbed props, required permission and limits #19879: the current PR reports MERGEABLE. No gratuitous rebase was performed; a real conflict after that PR lands will be handled separately.
  • Marked dispositions: each lane needs a marker at the very start of its own comment (_review_contract.py anchors it with \A). This combined reply is the point-by-point record, not a claim that both machine-readable lane dispositions are cleared.

One new commit, two total; no merge, auto-merge, review dismissal, or finding override.

@buluoray
buluoray requested a review from iamwhatever October 14, 2026 15:23
@github-actions github-actions Bot added readiness: checking Automated validation is still running and removed readiness: action required A blocking check or review needs attention labels Oct 14, 2026
@buluoray

Copy link
Copy Markdown
Contributor Author

🤖 Kiro Crew Auto-Pipeline [operator: buluoray#19887]:

  • Item 1 — symptom-level sibling fallbacks: fixed tracking; sibling code accepted-and-deferred.

Pattern harvest now names both sites and links their separately actionable tier:T1 issues: #20348 (CrewChatWindow header) and #20349 (SessionTabStrip labels).
Both fallbacks remain on main. This PR changes ChatEmbed only; no shared fix is claimed. Each issue records reproduction and regression-test acceptance criteria.
These issues are intentionally unassigned; no owner or due-date commitment is being claimed.

@buluoray

buluoray commented Oct 14, 2026 •

Copy link
Copy Markdown
Contributor Author

🤖 Kiro Crew Auto-Pipeline [operator: buluoray#19887]:

  • Hover tooltip / data-slot-key or conditional tooltip suggestion — retained.

The slot key remains the documented hover title for debugging on both titled and untitled headers. It is visible to anyone who hovers, not only developers; the claim is that it no longer replaces the visible label, not that it is hidden everywhere. The human review calls conditional hover optional. This PR retains that trade-off and makes no code change.

@buluoray

Copy link
Copy Markdown
Contributor Author

🤖 Kiro Crew Auto-Pipeline [operator: buluoray#19887]:

@iamwhatever @CrysisDeu — the separate marker-first dispositions are now posted: First Principles and UX.

  • Both requested sibling issues are open and linked in the description: fix: hide untitled slot ids in the remote crew chat header #20348 and fix: hide untitled slot ids in session tab labels #20349, tier:T1, unassigned.
  • Commit 1d8d35119e7ff493a7846ce092295975239cb25b already contains the fresh/resumed title-poll tests and owning-spec update. All 58 related tests pass, including shuffled order. Reverting exactly to {title || slotKey} makes each new case fail. TypeScript, ESLint and 53 flake-pattern checks pass.
  • The optional titled-hover and Hindi/Russian wording suggestions remain unchanged, with their reasons in the full point-by-point reply. The resumed-session question is covered by the new test and spec. The frozen Goal, Why and Not a goal remain unchanged.
  • The tests were already pushed once; the PR has two commits total. Review from iamwhatever was already re-requested once and remains requested. No duplicate push or review request was sent. New-head CI and AI review are still running.

@buluoray

Copy link
Copy Markdown
Contributor Author

🤖 Kiro Crew Auto-Pipeline [operator: buluoray#19887]:

  • Vendored SDK rebuild — no separate bundle is needed.

The vendor file is a re-export stub, not a compiled copy of ChatEmbed: website/public/vendor/kirocrew-app-sdk.mjs:2 reads window.__kirocrew_modules and :13 exports ChatEmbed from that object. website/src/app-sdk/shared-modules.ts:17 imports ./index and :42 registers it as @kirocrew/app-sdk; index.ts:483 exports the source ChatEmbed. The normal host build therefore supplies the changed component to external apps. No vendor generation or code push is required. This conclusion is source-traced; an external app was not browser-tested in this rebase.

@buluoray

Copy link
Copy Markdown
Contributor Author

🤖 Kiro Crew Auto-Pipeline [operator: buluoray#19887]:

  • Four additional sibling fallback sites — tracked outside this frozen scope.

Confirmed all four on main 0694e63: SessionRefStrip chip #20907, MembersPageProfile driving row #20908, CommandBar attention #20909 and recent #20910. One bug-kind tier:T1 issue per site, no assignee as instructed. Related Issues now links each; Pattern harvest labels the old 11-site list as historical, not exhaustive. CommandBar attention requires a pill status; recent excludes empty new slots and attention slots. No sibling code or frozen Goal/Why/Not-a-goal changed.

@buluoray

Copy link
Copy Markdown
Contributor Author

🤖 Kiro Crew Auto-Pipeline [operator: buluoray#19887]:

  • Raw slot-key hover tooltip — retained explicit trade-off.

The frozen change removes the raw key from visible header text, not from the existing diagnostic hover surface. The PR What changed and Backwards compatibility sections explicitly state that the slot key remains in title={slotKey}; the app-kit-platform spec and header test pin that behavior. The conductor ruled to preserve it and not push optional tooltip polish. This is not a claim that raw IDs are removed everywhere; no new user experience claim is added.

@github-actions github-actions Bot added readiness: passed Eligible automated validation passed for the current revision and removed readiness: checking Automated validation is still running labels Oct 15, 2026
iamwhatever
iamwhatever previously approved these changes Oct 15, 2026
@github-actions github-actions Bot added the merge conflict Branch has merge conflicts with its base — author must resolve before merge label Oct 15, 2026
An embedded chat with no title yet showed its internal slot key (for
example coder-abc123) as the header text. Fall back to a translated
"New session" label and keep the slot key only as the hover title.
@buluoray
buluoray force-pushed the fix/embed-untitled-header branch from 5cd8ab3 to b266889 Compare October 15, 2026 22:47
@github-actions github-actions Bot added readiness: checking Automated validation is still running and removed readiness: passed Eligible automated validation passed for the current revision merge conflict Branch has merge conflicts with its base — author must resolve before merge labels Oct 15, 2026
@buluoray

Copy link
Copy Markdown
Contributor Author

🤖 Kiro Crew Auto-Pipeline [operator: buluoray#19887]:

  • Vendored SDK rebuild — host export verified.

website/public/vendor/kirocrew-app-sdk.mjs reads window.__kirocrew_modules and re-exports ChatEmbed; it is not a compiled component copy. website/src/app-sdk/shared-modules.ts imports ./index and registers it under @kirocrew/app-sdk, whose index.ts exports ./ChatEmbed. The normal host build therefore supplies this source change. No separate vendor rebuild is required. This is source-traced; an external app was not browser-tested in this rebase.

@buluoray

Copy link
Copy Markdown
Contributor Author

🤖 Kiro Crew Auto-Pipeline [operator: buluoray#19887]:

  • Resumed untitled wording — retained scoped limit.

The frozen fallback is unchanged: an untitled resumed chat with messages also shows New session. The existing owner decision retains this wording here and tracks loading/age-aware titles in #20362. ChatEmbed.test.tsx parameterizes the title-poll test for a resumed session with Earlier turn, verifies New session initially, and verifies replacement by Trip planning on the same header node. That test passed at this head among 57 related tests. This proves the current behavior, not a fix for the deferred wording.

@buluoray

Copy link
Copy Markdown
Contributor Author

🤖 Kiro Crew Auto-Pipeline [operator: buluoray#19887]:

  • Resumed chat with messages — checked, wording remains deferred.

The resumed-session case in ChatEmbed.test.tsx supplies Earlier turn, asserts New session, then polls a real title and verifies the same mounted header becomes Trip planning. It passed in the 57-test related run at this head. The known misleading New session wording for resumed untitled chats is retained under the existing scope decision and remains tracked in #20362. No code change or broader UX claim is made.

@github-actions github-actions Bot added readiness: passed Eligible automated validation passed for the current revision and removed readiness: checking Automated validation is still running labels Oct 15, 2026
@buluoray
buluoray requested a review from iamwhatever October 15, 2026 23:40

@KiroCrewCaptain KiroCrewCaptain left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Tier 1 auto-approve: fix (18 files). Criteria: no conflict, no requested changes, security path denylist clean, design-doc gate clean, security checklist all-NO, no reviewer-lane findings, readiness signal fresh for this head SHA, declared category matches the diff. Category: an untitled ChatEmbed header shows a localized "New session" label instead of the internal slot key, which moves to the hover title; one new i18n string per locale plus tests. Spec files changed as a ride-along (a minority of the diff on both file count and changed lines), not reviewed as a design decision: docs/system-specs/modules/app-kit-platform.md.

@buluoray
buluoray merged commit 9eff131 into main Oct 16, 2026
103 of 104 checks passed
@buluoray
buluoray deleted the fix/embed-untitled-header branch October 16, 2026 00:05
@github-actions github-actions Bot removed the readiness: passed Eligible automated validation passed for the current revision label Oct 16, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants