Skip to content

feat(chat): offer resume on a signed-out row once kiro sign-in works - #35787

Open
bolichen97 wants to merge 1 commit into
mainfrom
fix/34690-resume-after-kiro-sign-in
Open

bolichen97 wants to merge 1 commit into
mainfrom
fix/34690-resume-after-kiro-sign-in

Conversation

@bolichen97

@bolichen97 bolichen97 commented Nov 12, 2026 •

Copy link
Copy Markdown
Contributor

Problem / Motivation

Goal: Once Kiro's own sign-in works again, a chat turn that failed with "Not signed in to Kiro" can be re-run from its own row in the same chat, without typing the message again.

A turn sent while Kiro is signed out ends on a row whose only button is Sign in and whose text says to send the message again. After the user signs in, the row still reads the same and still offers only Sign in, so the user types the first message again (older builds even said to start a new chat). The onboarding bug bash hit this on a first-run chat ("What can you do for me today?").

Why it matters

This is often the very first message a new Kiro Agent user sends. Losing it at that moment, and having to find and retype it after signing in, makes the first run feel broken.

Not a goal

  • Re-sending the turn on its own without a press: Resume runs a full agent turn, so it stays the user's choice.
  • Signing in from inside the chat row: the button still opens the Kiro sign-in card in Settings.
  • The launcher form of the row, whose fix is kiro-cli login: Kiro's own sign-in cannot clear it, so it keeps its button.
  • Chat panes, which do not read the sign-in state: their row is unchanged.

What changed (motivation → approach → change)

After a signed-out failure the row never offered Resume, because every auth_required row withholds it: a retry would hit the same wall. That stops being true once the sign-in works, but nothing looked.

The chat page reads Kiro's sign-in status (/api/kas-login, the query the sign-in card already uses) only while the chat's latest turn ended on a kas_signed_out row: a request sent after that row is the one Resume would rerun, so such a row keeps only Sign in and the page reads nothing for it. When the latest read says signed in, not expired and not lapsed, the row keeps its Sign in button and gains Resume beside it, worded "Signed in to Kiro — resume to send your message again.". Pressing Resume re-runs the turn in the same chat through the existing Continue path. The Sign in button stays because the status can read signed in while Kiro still refuses the token (a revoked session), and the read is keyed by the row, so a status fetched before the row appeared never counts. Embed and popout, which have no Sign in button, show Resume alone.

The row re-reads the status every minute while it stands, so a popout signed in from another window gains Resume too. A read that fails withdraws Resume, even after an earlier read said signed in, and the next good read brings it back. The popout and embed frames (/popout/*, /embed/*), which have no app-level sign-in notice, show that failure on the row itself: "We could not check sign-in status" with the server's sentence, through ErrorNotice and with no agent hand-off, since the agent runs as the same Kiro sign-in the row says failed. The dashboard window leaves it to the app-level sign-in notice, which reads the same endpoint, so the row adds no second notice there; that includes a chat embedded in a dashboard page, such as an artifact's companion chat. A read that fails while the dashboard is offline (a gateway restart) is not shown as a sign-in problem: useOutageFailure leaves it to the offline state, as the app-level notice does.

Piece What it does
kiroSignedIn in page/transcriptRendering.tsx true only on a successful latest read that says the sign-in works; the read repeats every KIRO_SIGN_IN_NOTICE_POLL_MS
ErrorCard signInWorks draws Resume in the sign-in card's own button row, so the card keeps one shape
ErrorCard signInCheckError the failed status read, shown under the row's text in a popout or embed frame, told apart by the window path as AgentSwitchNoticeBanner does
useOutageFailure on the row's read drops a failure taken while the dashboard is offline, so a gateway restart puts no sign-in notice on the row
signedInNow in kiroSignInState.ts signed in, not expired, not lapsed: usable alone reads an expired token as fine when a refresh token is stored
isKasSignedOut in gatewayNotice.ts tells this row from the launcher form, whose fix is kiro-cli login
pages.chat.errorCard.retry_kas_signed_in the new string, in every catalog under website/src/i18n/locales/

kas-auth.md records the behavior.

Backwards compatibility

Compatible: two optional ErrorCard props, two optional renderer options and one new catalog key; every other row, and this row while signed out with a working status read, renders as before.

Removes nothing: the Sign in button stays on every such row; Resume is added beside it.

Tests

  • website/src/test/transcriptRenderers.test.tsx: a kas_signed_out row offers no Resume while signed out, and once kiroSignedIn is true it offers Resume with the new wording and still passes the Sign in handler; the launcher form (kiro-cli login) offers no Resume even when signed in, and neither does a row a later request follows; signedInNow reads an expired token, or a refused renewal, as not signed in even when usable is true; on a frame with no Sign in button, a failed status read renders on the row with its title and the server's sentence and no "Ask the agent".
  • website/src/test/ErrorCard.test.tsx: with signInWorks, the sign-in card shows the resume wording and both buttons, and each calls its own handler.
  • website/src/pages/ChatPage.transcriptRendering.test.tsx, through the page's own hook: in an embed frame, a signed-in read draws Resume and a refetch that then fails (a 500 token_store_failed) withdraws it and draws the failure notice; on the dashboard a failed read keeps Sign in and draws neither Resume nor a notice on the row; in a chat embedded in a dashboard page (/artifacts/doc-1) a failed read draws neither a notice nor Resume on the row; in an embed frame, a read that fails (Failed to fetch) while the gateway is disconnected draws no notice and no Resume, and the hook's own re-read once it is back brings Resume; in a frame whose first read says signed out, only the clock advancing by the poll interval (fake timers, no manual refetch) brings Resume once the status says signed in; a row a later request follows draws no Resume and makes no status read. Each fails on the code without its change.

With transcriptRenderers.tsx or ErrorCard.tsx reverted, the Resume case in its spec fails; with the three chat files reverted, the failed-read case fails (no notice on the row); with the frame decided by the embedded prop instead of the window path, the companion-chat case fails (the row notice is drawn). The launcher case holds on both sides by design: it pins that the launcher form is not widened. Without the latest-turn check, the two later-request cases fail (Resume offered, and the status read made). The three spec files passed 198 of 198 in three runs and in one shuffled run.

Manual verification

On Linux, an isolated gateway (a temporary data home on 127.0.0.1, the default Kiro Agent backend, no Kiro sign-in) served the dashboard built from this branch, driven with Playwright in Chromium, with /api/kas-login answered by the test. The first turn was typed and failed for real; that host's engine words the failure in a sentence the gateway does not file as signed out, so the one error row was rewritten to the shape the gateway stores for a Kiro signed-out failure before the page loaded.

  • On main, with the read saying signed in, the dashboard row offers only Sign in and the popout row offers nothing.
  • On this branch the row reads "Signed in to Kiro — resume to send your message again." with Sign in and Resume; pressing Resume with an empty composer sent POST /api/chat/slots/<slot>/continue, and the same chat gained [Continue — requested by the user] and a new agent attempt.
  • In a popout frame: a signed-in read shows Resume; a failed focus refetch withdraws it and shows the notice; the next good read brings Resume back. A popout whose first read said signed out gained Resume about a minute after the status switched to signed in, with no reload or focus change (without the re-read: none within 70 s).
  • On the dashboard a failed read keeps Sign in, shows no Resume and no notice on the row, and the app-level banner at the top says "We could not check sign-in status"; once the read worked, the minute re-read brought Resume back and the banner cleared.
  • In an artifact's companion chat (/artifacts/<slug>, the agent chat beside the document), a failed read shows the app-level banner only: the row keeps its signed-out text with no notice and no Resume (on the previous revision the row drew a second "We could not check sign-in status"); once the read said signed in, the banner cleared and the row gained Resume. The same chat opened as a popout still draws the notice on its row.
  • In a popout with the gateway stopped, four failed re-reads over about two minutes drew no notice on the row (on the previous revision, "We could not check sign-in status / Failed to fetch" appeared beside the offline state); after the gateway came back the row was in its signed-out shape with no notice. With the gateway up, a 500 still draws the notice on the popout row.
  • An ordinary (not sign-in) error row still offers Resume whatever the sign-in read says, as on main.
  • A chat whose signed-out row A is followed by an unanswered request B, with the read saying signed in: row A keeps its signed-out text and Sign in, offers no Resume and draws no notice, and the page makes one /api/kas-login read, the app-level notice's (a page with no chat row makes the same one). Without the latest-turn check, row A offered Resume and pressing it queued a recovery turn told to finish the user's most recent request, which is B. A chat that ends on row A still offers Resume, and the press sent POST /api/chat/slots/<slot>/continue (200).

From the smoke run on this branch (popout frame, then the Resume press on the dashboard):

STATE {"text":"Signed in to Kiro — resume to send your message again. Resume","resume":1,"signIn":0,"checkFailed":0}
kas-login read #2 -> fail
STATE-AFTER-FAILED-REFETCH {"text":"Not signed in to Kiro. … We could not check sign-in status Could not read the stored sign-in.","resume":0,"signIn":0,"checkFailed":1}
kas-login read #4 -> in
STATE-AFTER-GOOD-REFETCH {"text":"Signed in to Kiro — resume to send your message again. Resume","resume":1,"signIn":0,"checkFailed":0}
composer text before press: ""
POSTS ["POST /api/chat/slots/<slot>/continue"]

Then the re-read in a popout:

R1 popout: [+6.7s] route switched to signed in -> [+62.5s] RESUME APPEARED {resume:1, text:'Signed in to Kiro — resume to send your message again. Resume'}

And from the smoke run on the current revision (popout with the gateway stopped and started again; dashboard with a failed read, then signed in):

while offline: row polls failed = 4, max sign-in-check notice count seen = 0, composer placeholder = "Gateway offline — message will not send"
after restart: notice = 0 resume button = 0 row text = "Not signed in to Kiro. Sign in from Settings → Agent Harness → Kiro Agent → Kiro sign-in, then send your message again."
dashboard after failed read: {"banner":"We could not check sign-in status ...","rowNotice":"absent","signInButton":1,"resume":0}
dashboard after signed-in read: {"banner":"absent","rowNotice":"absent","signInButton":1,"resume":1,"rowText":"Signed in to Kiro — resume to send your message again. ..."}
popout after failed read: {"banner":"absent","rowNotice":"We could not check sign-in status / Could not read the stored sign-in.","resume":0}

A signed-out row followed by a later request, signed-in read, then a chat ending on that row:

row A text: "Not signed in to Kiro. Sign in below, then send your message again.\nSign in to Kiro"
sign-in button: 1   error-card-continue: 0   sign-in-check-failed notice: 0
GET /api/kas-login reads: 1   (control page /settings: 1)
continue POSTs: []
chat ends on row A: row A text: "Signed in to Kiro — resume to send your message again.\nSign in to Kiro\nResume"
POST /api/chat/slots/<slot>/continue -> 200

And the artifact companion chat (banner is the app-level notice, rowNotice the row's own):

previous revision, failed read: {"banner":true,"rowNotice":true,"rowResume":false,"signedOutText":true}
failed read (en dark, en light, ja dark): {"banner":true,"rowNotice":false,"rowResume":false,"signedOutText":true}
after a good read: {"banner":false,"rowNotice":false,"rowResume":true,"signedOutText":false}
same chat as a popout, failed read: {"banner":false,"rowNotice":true,"rowResume":false}

Screenshots / video

Dashboard: signed out (top) and signed in (bottom), English, dark:

dashboard row signed out and signed in, English dark

English, light:

dashboard row signed out and signed in, English light

Japanese, dark:

dashboard row signed out and signed in, Japanese dark

Popout frame, signed in: Resume and no Sign in button, English, dark:

popout row signed in, English dark

English, light:

popout row signed in, English light

Japanese, dark:

popout row signed in, Japanese dark

Popout frame, top to bottom: signed out with a working read; the first read failed; a signed-in read followed by a failed refetch (Resume withdrawn), English, dark:

popout row with a working read, a failed read and a failed refetch, English dark

Japanese, dark:

popout row with a working read, a failed read and a failed refetch, Japanese dark

Dashboard, after the sign-in read failed: the app-level banner at the top, the row with Sign in kept, no Resume and no notice of its own, English, dark:

dashboard after a failed sign-in read, English dark

English, light:

dashboard after a failed sign-in read, English light

Japanese, dark:

dashboard after a failed sign-in read, Japanese dark

An artifact's companion chat after the sign-in read failed: the app-level banner at the top and the row with its signed-out text, no notice and no Resume of its own, English, dark:

artifact companion chat after a failed sign-in read, English dark

English, light:

artifact companion chat after a failed sign-in read, English light

Japanese, dark:

artifact companion chat after a failed sign-in read, Japanese dark

The same companion chat on the previous revision, which drew a second "We could not check sign-in status" on the row under the banner:

artifact companion chat after a failed sign-in read on the previous revision, English dark

Popout frame with the gateway stopped: the offline composer and the row with no sign-in notice, English, dark:

popout row during a gateway outage, English dark

The same outage on the previous revision, which drew a false "We could not check sign-in status / Failed to fetch" on the row:

popout row during a gateway outage on the previous revision, English dark

A signed-out row followed by a later request, with the read saying signed in. Without the latest-turn check, the row offers Resume, which would rerun the later request, English, dark:

signed-out row followed by a later request, without the latest-turn check, English dark

With it, the row keeps Sign in and offers no Resume, English, dark:

signed-out row followed by a later request, English dark

English, light:

signed-out row followed by a later request, English light

Related Issues

Closes #34690

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

@bolichen97
bolichen97 requested review from a team and KiroCrewCaptain as code owners November 12, 2026 08:34
@github-actions github-actions Bot added the readiness: checking Automated validation is still running label Nov 12, 2026
@github-actions

github-actions Bot commented Nov 12, 2026 •

Copy link
Copy Markdown
Contributor

Design Review (Opus 5.5) — ✅ PASS

Design-level review of 28a62d416c81a36f9393e1321248ac615199e3d2 — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

Design-Verdict: PASS

Nothing blocks: Resume shows up only on the newest kas_signed_out row, after a fresh status read keyed to that row, and it reuses the existing Continue path.

Notes

  • signedInNow is stricter than !signInLapsed, so the two can now disagree. If the Settings sign-in card and the app-level notice keep using the looser check, they can call an expired token "fine" while this row withholds Resume. Moving those readers to signedInNow would be a follow-up, not part of this PR.

[DESIGN-REVIEWED] 28a62d4

Description read

[DESCRIPTION-READ] 3cc7c1913b64aaf60b8fd8565d7f30b6aa15014878c556b9b5ae2fe91dd87070

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 Nov 12, 2026 •

Copy link
Copy Markdown
Contributor

First Principles Review (Opus 5.5) — 🔴 BLOCK

Premise-level review of 28a62d416c81a36f9393e1321248ac615199e3d2 — 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 does not.

First-Principles-Verdict: BLOCK

The new Resume button on the signed-out row repeats a Resume the composer already shows for that same row. It sends the same /continue press, and the description never mentions the existing one.

Notes

  • Item 4 — oversized: the failed sign-in read is shown on the row in popout/embed (signInCheckNotice, useOutageFailure, the framed path test). This is a second error surface the Goal never asked for. Subtraction: delete signInCheckError/kiroSignInCheckError and the framed/kasStatusOutage wiring.
  • Item 3 — oversized: the one-minute refetchInterval poll serves only the "signed in from another window" case. Subtraction: drop refetchInterval; a sign-in through the ['kas-login'] card still triggers a re-read.

What this change ships

Inventory (6 items) — 2 justified

Intent: let a turn that failed with "Not signed in to Kiro" run again after signing in, without retyping it. ADDITION.

  1. Signed-out row gains a Resume button once the sign-in reads as working — duplicate of ChatInput.tsx composer Resume (resumeAfterSignIn)
  2. Row wording changes to "Signed in to Kiro — resume to send your message again." — justified
  3. Chat page reads /api/kas-login every minute while such a row stands — oversized
  4. Popout/embed row shows "We could not check sign-in status" when the read fails — oversized
  5. signedInNow helper treats expired tokens as signed out — justified
  6. New retry_kas_signed_in string in 13 locale files — rides along

Blockers

Row Resume duplicates the composer's sign-in Resume — duplicate of website/src/components/ChatInput.tsx
Trigger: authRequiredIsNewest (transcriptRenderers.tsx:381) says: "The card withholds its own Continue, but the composer keeps Resume: once the sign-in lands, the same press works… it is the one-press way back." I grepped authRequiredIsNewest and counted 3 composer callers: turnRecovery.ts, useChatPaneContinue.ts and CrewChatWindow.tsx. The composer's hint already reads "then press Resume" (turn_interrupted_sign_in_first). The description says "still offers only Sign in, so the user types the first message again", which leaves this out. It also reverses the "card withholds" decision without saying so.
Outcome: two Resume controls for one turn, each with its own rules for when it appears, kept in step forever. The row's copy also contradicts the composer's.
Subtraction: delete the row-level onContinue/signInWorks branch and the kas-login query in transcriptRendering.tsx. Keep the composer's Resume.
Clears when: the row Resume is removed, or the description shows why the composer's existing Resume fails the Goal on this row.

[FIRST-PRINCIPLES-REVIEWED] 28a62d4

Description read

[DESCRIPTION-READ] 31ddd4a61e1ef1e3dad98ac945d437fbbaf7e25c1e63a08a62297d67978469db

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 Nov 12, 2026 •

Copy link
Copy Markdown
Contributor

UX Review (Opus 5.5) — ✅ PASS

UX-level review of 28a62d416c81a36f9393e1321248ac615199e3d2 — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

UX-Verdict: PASS

Nothing blocks. The blind reader understood Resume on the main path (shots 01 and 04) and would press it. The confusions below are notes, not defects that hold the merge.

Notes

  • When the row reads "Signed in to Kiro — resume…", the button beside Resume still says "Sign in to Kiro" (shots 01–03, 18). The blind reader asked: "Would it log me out, or make me log in twice?" Shots 04–06 have no such button, and the reader couldn't tell why the two versions differ. While signInWorks is true, relabel that button "Kiro sign-in settings", so it names where it goes instead of asking for a sign-in that already worked.
  • Shot 18 shows a follow-up message (Message B) sent while the turn was stopped, sitting below the Resume row. It looks exactly like a normal message, so the reader had "no idea" whether Resume resends A or both, and would hesitate. Either mark that follow-up message as one that will ride along with the resumed turn, or say in the row's copy what Resume will resend.
  • The good-news row is still drawn as a red error (the reader: "I can't tell if something is still wrong"). Once signInWorks is true, give the row a neutral or info style, so the colour stops saying something is wrong after sign-in works.

[UX-REVIEWED] 28a62d4

Description read

[DESCRIPTION-READ] 630d2bbca46eac867705558d51c0af7f51e0c223f8a7ee11822c95d17faaeb02

That sha256 names everything this verdict read: the PR description, plus the 23 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 Nov 12, 2026 •

Copy link
Copy Markdown
Contributor

Opus 5.5 Review — ✅ no blocking findings

Reviewed 28a62d416c81a36f9393e1321248ac615199e3d2 — this comment is updated in place on each push.

Review details

No findings.

[OPUS-REVIEWED] 28a62d4

Verdict parsed from the review's SHA-scoped output markers for commit 28a62d416c81a36f9393e1321248ac615199e3d2.

False positive or not applicable? A repository writer can comment:
/ai-review override fable 28a62d416c81a36f9393e1321248ac615199e3d2 false-positive: <reason, 20 to 500 characters>

@github-actions

github-actions Bot commented Nov 12, 2026 •

Copy link
Copy Markdown
Contributor

GPT 6.1 Review — ✅ no blocking findings

GPT 6.1 completed its review of 28a62d416c81a36f9393e1321248ac615199e3d2 and found no blocking issues.

This comment is updated in place on each push.

Review details

No findings.
[GPT-REVIEWED] 28a62d4

False positive or not applicable? A repository writer can comment:
/ai-review override gpt 28a62d416c81a36f9393e1321248ac615199e3d2 false-positive: <reason, 20 to 500 characters>

@bolichen97
bolichen97 enabled auto-merge (squash) November 12, 2026 08:49
@github-actions github-actions Bot added readiness: action required A blocking check or review needs attention readiness: checking Automated validation is still running and removed readiness: checking Automated validation is still running readiness: action required A blocking check or review needs attention labels Nov 12, 2026
@bolichen97
bolichen97 force-pushed the fix/34690-resume-after-kiro-sign-in branch from c86caa1 to bff2b58 Compare November 12, 2026 20:49
@bolichen97

bolichen97 commented Nov 12, 2026 •

Copy link
Copy Markdown
Contributor Author
  • Sign-in check failures are silent in embed and popout span=ee96865419be

fixed in bff2b58: the failed /api/kas-login read now reaches the row in embed and popout, rendered through ErrorNotice (inline, title "We could not check sign-in status" plus the server's sentence) on the newest kas_signed_out row.
Hand-off: noHandoffReason, the rule's fourth state: on KAS the agent runs as the stored Kiro sign-in this row reports failed, so an Ask-the-agent chat would fail the same way.
Dashboard: unchanged, because the app-level KiroSignInNotice (mounted above every page) reads the same endpoint and renders its failure through ErrorNotice.
Test: transcriptRenderers.test.tsx "shows a failed sign-in check on the row where no sign-in button or app notice reports it", red with the three chat files at the old head, green with the fix.

[operator: bolichen97]

@github-actions github-actions Bot added readiness: action required A blocking check or review needs attention merge conflict Branch has merge conflicts with its base — author must resolve before merge and removed readiness: checking Automated validation is still running labels Nov 12, 2026
@bolichen97
bolichen97 force-pushed the fix/34690-resume-after-kiro-sign-in branch from bff2b58 to a5a6b74 Compare November 13, 2026 07:24
@bolichen97

Copy link
Copy Markdown
Contributor Author

Rebased: main moved in website/src/pages/chat/page/transcriptRendering.tsx since this head went green (3eff86a -> 00c5418); CI re-running.

The branch conflicted with main in that file (the renderer block's dependency list); resolved as the union of both sides.

[operator: bolichen97]

@bolichen97

Copy link
Copy Markdown
Contributor Author

self-added: yes
mechanism: the row's own /api/kas-login read decides Resume and reports its own failure

  • Failed status refetches disappear while Resume is offered span=76afe16638cc

fixed in a5a6b74: kiroSignedIn now needs a successful latest read (kasStatusError === null), so a failed refetch withdraws Resume instead of trusting React Query's kept data, and the row never says "Signed in" over a failure.
Class: the failure and Resume are now mutually exclusive on that row; the one ErrorNotice block is built once and drawn on the dashboard's sign-in card and on the frames' row, so the read's failure shows on every surface.
The row also re-reads every KIRO_SIGN_IN_NOTICE_POLL_MS, so a frame signed in from another window gains Resume and a failed read's notice clears on recovery.
Tests: ChatPage.transcriptRendering.test.tsx (failed refetch withdraws Resume and shows the notice; dashboard failed read beside Sign in; the poll brings Resume with only the clock advancing), each red on the prior code.

[operator: bolichen97]

@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 Nov 13, 2026
@github-actions github-actions Bot added readiness: action required A blocking check or review needs attention and removed merge conflict Branch has merge conflicts with its base — author must resolve before merge readiness: checking Automated validation is still running labels Nov 13, 2026
@bolichen97
bolichen97 force-pushed the fix/34690-resume-after-kiro-sign-in branch from a5a6b74 to 8272e81 Compare November 13, 2026 12:17
@bolichen97

Copy link
Copy Markdown
Contributor Author

Rebased: main moved in src/kiro_crew/cli_commands_memory.py since this head's CI ran (00c5418 -> 6157e13); CI re-running.

This head's Backend Lint & Type Check (and the Coverage Gate behind it) failed on a mypy error in that file, which this PR does not touch; main fixed it in #36457, so the branch now sits on a base that carries the fix. The same push carries the fix for the Opus finding (disposition below).

[operator: bolichen97]

@bolichen97

Copy link
Copy Markdown
Contributor Author

self-added: yes
mechanism: the row's own /api/kas-login read reports its own failure on the row
resolution: narrowed -- the row's failed-read notice is back to embed and popout only, and outage failures are dropped there

  • The signed-out row's sign-in read shows a false notice during a gateway outage span=dc056ad837e7

fixed in 8272e81: the row's read now goes through useOutageFailure, as KiroSignInNotice does, so a failure taken while the dashboard is offline is left to the offline state; any failing read still withdraws Resume (kasStatusError === null in kiroSignedIn).
Class: the row's failed-read notice is narrowed back to embed and popout, where no KiroSignInNotice is mounted; on the dashboard that banner already reports a failed read of the same endpoint, outage-filtered, so each failure has one error surface.
Pinned by the embed outage case and the dashboard failed-read case in ChatPage.transcriptRendering.test.tsx, both red on the previous head; smoke: four failed re-reads over two minutes with the gateway stopped drew no row notice.

[operator: bolichen97]

@github-actions github-actions Bot added readiness: checking Automated validation is still running readiness: action required A blocking check or review needs attention and removed readiness: action required A blocking check or review needs attention readiness: checking Automated validation is still running labels Nov 13, 2026
@bolichen97

Copy link
Copy Markdown
Contributor Author

[operator: bolichen97] Backend Tests (3.12, 7) is red on head 8272e81 because six built-in app tests hit the 120 s pytest timeout: four in apps/builtins/design_critique/tests/test_ssrf_guard_fail_closed.py and two in apps/builtins/issue_radar/tests/test_write_and_ai.py (job log). Coverage Gate failed closed only because that shard did. Every other check, and every review lane, is green on this head.

This PR changes only website/ files and one spec doc, nothing a backend test imports. The same shard is red on main (run 43458639790, lost runner) and on many unrelated PRs. It is tracked in #36261, #36521 and #36515, and the fixes are open in #36653, #36648 and #36655.

So this red is main's, not this PR's: no rerun, no push and no override for it. This PR waits for those fixes, then rebases onto the one that lands.

@KiroCrewCaptain

Copy link
Copy Markdown
Collaborator

Code change approver recommendation: Approve

Why: the decider estimates a 12% chance that this change introduces a defect, within the 12% limit for approving on its own, and every check passes.

This frontend-only change adds a Resume button to the newest interrupted kas_signed_out error row once a status read taken after the row appeared says Kiro sign-in works (signedInNow). The Sign in button stays. The read is keyed by the row and sits under the shared ['kas-login'] prefix, so the sign-in card's invalidation reaches it, and every other reader uses the same fetcher. The read repeats on the existing notice interval. A failed read withdraws Resume. In embed and popout the failure shows on the row, except during a gateway outage, which useOutageFailure leaves to the offline state. The launcher form and chat panes are unchanged, and the existing Continue path does the re-run. Tests cover the gate, the launcher case, the expiry and refused-renewal edge cases, the ErrorCard buttons and the hook's poll, failure and outage paths. The failing backend shard is timeouts in unrelated Python tests, and the Coverage Gate failure most likely follows from it.

An automated recommendation, not a review: a maintainer still decides. It approves on its own only when the estimated chance of a defect is at most 12% and the change has at most 400 changed lines. Decider v3c (217fe607) · commit 8272e814e3 · 359 changed lines. If this recommendation is wrong, reply here: replies are how the approver is evaluated.

@buluoray

Copy link
Copy Markdown
Contributor

Review passed on this head. Please rebase onto the latest main so CI can run fresh; the approval will follow once the rebased head is green.

After a turn failed because Kiro was not signed in, the chat row only
offered the sign-in, even once the sign-in had landed, so the failed
message had to be typed again. The chat page now reads the Kiro sign-in
status while its turn ended on that row; once the sign-in works, the
newest such row, when no later request follows it, offers Resume beside
its sign-in button, which reruns the turn in the same chat. The button
stays, since the status can read signed in while the engine still
refuses the token. The row re-reads the status every minute while it
stands, and a failed read withdraws Resume rather than trusting the
earlier answer. The popout and embed frames, which carry no app-level
sign-in notice, show that failure on the row, except one taken while the
dashboard is offline, which the offline state already reports; a chat
embedded in a dashboard page leaves it to that notice.

Closes #34690
@bolichen97
bolichen97 force-pushed the fix/34690-resume-after-kiro-sign-in branch from 8272e81 to 28a62d4 Compare November 17, 2026 19:13
@bolichen97

Copy link
Copy Markdown
Contributor Author

Rebased: main moved in website/src/i18n/locales/*.json (the 13 locale files this PR edits), website/src/pages/chat/ErrorCard.tsx, website/src/pages/chat/gatewayNotice.ts, website/src/pages/chat/transcriptRenderers.tsx, website/src/pages/chat/page/transcriptRendering.tsx, website/src/test/transcriptRenderers.test.tsx and docs/system-specs/modules/kas-auth.md since this head went green (6157e13 -> 6c0cc77); CI re-running.

This is the rebase asked for above. The only CI reds on 8272e81 were Backend Tests (3.12, 7) (built-in app tests hitting the 120 s timeout, #36261, now closed) and Coverage Gate failing closed behind it. That red is main's and its fixes are on main (#36731, #36648), so this rebase picks them up. Main's edits in these files sit next to this PR's lines at most; the rebase changed only context lines.

The push also carries the fix for the GPT finding on 8272e81 (span e704535aaf87): a signed-out row that a later request follows offers no Resume and gets no sign-in read, because Resume would rerun that later request rather than the row's own. A renderer test and a page-hook test cover it, and the description has three new screenshots of that state.

[operator: bolichen97]

@bolichen97

Copy link
Copy Markdown
Contributor Author

self-added: no

  • Resume can target a different turn than its sign-in error row span=e704535aaf87

fixed in 28a62d4: a signed-out row offers Resume, and the page reads the sign-in status for it, only while authRequiredIsNewest holds (nothing after the row opens a turn), the shape the file already uses for contextOverflowIsNewest and imageRepairIsNewest.
Class: all three sites take the check: onContinue and signInCheckError in transcriptRenderers.tsx, and signedOutRow (the status read) in page/transcriptRendering.tsx; so [user A, signed-out A, user B] keeps only Sign in on A and makes no row read.
A chat that ends on the signed-out row keeps Resume and its read, so the goal holds.
Tests: a transcriptRenderers.test.tsx assertion and a ChatPage.transcriptRendering.test.tsx case, both red without the check; smoke on a real gateway: no Resume on A with B after it, Resume and POST .../continue 200 when the chat ends on A.

[operator: bolichen97]

@github-actions github-actions Bot added readiness: checking Automated validation is still running readiness: action required A blocking check or review needs attention merge conflict Branch has merge conflicts with its base — author must resolve before merge and removed readiness: action required A blocking check or review needs attention readiness: checking Automated validation is still running labels Nov 17, 2026
@bolichen97

Copy link
Copy Markdown
Contributor Author

Re-sync budget reset

This PR is inside its 48-hour watch window and GitHub now reports it as conflicting with main, so it needs one more re-sync before it can merge. Re-syncs so far: 3 of 3.

[operator: bolichen97]

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

merge conflict Branch has merge conflicts with its base — author must resolve before merge readiness: action required A blocking check or review needs attention

Projects

None yet

Development

Successfully merging this pull request may close these issues.

After 'start a new chat', the original message is not re-sent automatically after sign-in

3 participants