Skip to content

feat(followup): opt-in one-click send for follow-up Start, with a prompt preview - #25926

Merged
bolichen97 merged 1 commit into
mainfrom
pearcekieser/feat-followup-start-sends
Nov 10, 2026
Merged

bolichen97 merged 1 commit into
mainfrom
pearcekieser/feat-followup-start-sends

Conversation

@Pearcekieser

@Pearcekieser Pearcekieser commented Oct 29, 2026 •

Copy link
Copy Markdown
Contributor

Problem / Motivation

Goal: A person who turns it on can start a follow-up suggestion with one click, so its prompt goes to the new session as the first message. By default, Start keeps pre-filling the new session as it does today.

Today "Start in new worktree" and "Start in new chat" open the new session with the prompt in its message box, so every start needs a second press of Send. Some people want the click on Start to be the decision to run it. A click that sends agent-written text changes consent, so it is an opt-in and the default does not change.

The card also has two problems in both modes. The button says "worktree", which a first-time reader does not understand. The card shows only the agent's title and description, not the text that runs.

Why it matters

A follow-up exists to hand work to a new session in one step. With the pre-fill, a person who switches away before pressing Send leaves an empty session behind with the prompt in its draft. The people who want one step can now have it, and nobody else gets a click that runs agent text.

Not a goal

  • Changing the default. Start pre-fills unless the person turns on the setting.
  • Removing the switch to the new session. Running a follow-up in the background is a separate change.
  • Changing "Add to this session", which keeps pre-filling the current message box.
  • Persisting the card server-side.

What changed (motivation → approach → change)

The setting is the browser-local key mc-followup-start-sends (website/src/pages/chat/followupStartSends.ts), shown as "Send follow-ups with one click" under Settings, Chat, Composer (website/src/pages/settings/ChatPanel.tsx). It is read through usePersistedBool, so only the stored string '1' is on. It is not a ChatConfig field and not in DURABLE_PREF_KEYS, because the uiPrefs host backup is agent-writable (the mc-yolo-ack rule).

ComposerDockView.tsx reads it and website/src/components/FollowUpCard.tsx takes it as startSends. That one value picks the label (Start or Run), the send glyph and the footer, and goes to Start as mode: 'send' | 'prefill'. Each one-click footer opens with why it sends and a link to turn that off. In openFollowupSession (website/src/pages/useChatPageSlotSwitches.ts) only an explicit 'send' sends.

The pre-fill path is main's code and keeps main's worktree messages. The one-click path uses new *_sent_* keys that point at the sidebar.

The one-click path sends item.prompt verbatim through the composer's send() to the new slot by key, before the switch. A new send() callback in composerSend.ts reports a refusal or a lost answer. The RFC's failure table applies:

  • A refused send closes the empty session and shows the reason and a retry on the item.
  • If that session cannot be closed, the item leaves the card and one page notice points at the sidebar, with no retry (a retry on the worktree route gets a 409). That close is main's deleteSlot({ key, discard: true }), which refuses unless the session is still empty.
  • A lost answer counts as unconfirmed. The item leaves the card, nothing resends it, and the notice says "this session" when the person is on it. recordSendAttempt keeps the prompt on ↑.
  • An accepted send removes only its item, then switches, unless the person moved meanwhile (captureForkOrigin, forkMayOpen).

After the person leaves the Chat route, failures go to the shell banner.

Each item shows the first three lines of its prompt, with "Show prompt" for the rest. The worktree button reads "Start in a separate copy", its tooltip names the git worktree, and both of its footers say what the copy is. Every footer says "message box", and every worktree message says "copy".

The tool description in schemas.py, the model note in session_directive_apply_slot.py, followup-suggestions.md, the spec paragraph in learn-cron-dashboard.md and the trust-model doc describe both modes. rfc-followup-start-sends.md and its index row record the 2026-11-05 amendment. New strings are translated in all 12 catalogs, and en-XA.json, the settings registry and the UI index are regenerated.

Security note. With the setting off, nothing runs until the person presses send, as on main. With it on, one click sends agent-written text as a user turn in a new session. Before that click the person sees the title, the description and the first three lines of the exact prompt, as plain text, with the rest one toggle away. The prompt is still redacted server-side when the card is posted. The setting is off by default, is never restored from the host backup, and is not writable by a change card.

Backwards compatibility

Compatible. The default behaviour of Start is unchanged, and the suggest_followup tool's arguments and result are unchanged.

The two footer keys both_actions_pre_fill_the_composer_nothing_is_se and new_chat_footer_same_project_directory are replaced by footer_both_prefill and footer_both_prefill_project_directory, which say "message box". FollowUpCard.tsx was their only reader.

Tests

  • ChatPageFollowup.test.tsx, with the real card and composer: by default Start creates the worktree, scopes the session and puts the prompt in its message box with no send, and the new-chat route does the same. With the setting on, Start sends the prompt once to the new slot and leaves every message box empty. A double click sends once. A refused send shows on the item with no second notice.
  • ChatPageW3Coverage.test.tsx: the pre-fill block covers the prefill channel, a route called with no mode, no key, a switch that fails with the item kept, and focus that never arrives. The one-click block covers send before switch, the refused, kept and close-failed rows, the withdrawn close notice, both unconfirmed wordings, ↑ recall, and a person who moves during the start.
  • FollowUpCard.test.tsx: the default footer and no send glyph, the one-click footers and glyph, the mode passed to Start, the plain label and the tooltip, the three-line preview and Show prompt.
  • SettingsChatPanelCoverage.test.tsx: the toggle is off by default, writes '1' and '0' under its own key, never touches mc-chat-config, and the key is not in DURABLE_PREF_KEYS. ChatPageFollowup.test.tsx also checks that a stored 'true', or a followupStartSends field in the chat-config blob, keeps Start pre-filling.
  • switchSlotCallsiteClassification.test.ts classifies the second switchSlot site.

Proof: not provable -- frontend test

Manual verification

I rendered the built SPA with every API call stubbed and dummy data. With the setting off, Start in a separate copy opened the new session with the prompt in its message box and sent no request to /api/chat. With the setting on, one POST /api/chat carried the prompt to the new session.

Screenshots / video

The images are in temp-screenshots/followup-start-sends/.

Before, the card footer says both actions pre-fill:

before card

Before, Start opens the new session with the prompt pre-filled:

before start

After, by default: the plain label, no send glyph, the first three lines of each prompt, and a footer that says nothing is sent:

after card

After, the first item's prompt expanded:

after prompt shown

After, by default, Start opens the new session with the prompt in its message box:

after start

After, by default, in a project that is not a git repository:

after no-git card

The setting, off and on:

setting off

setting on

With the setting on, Start carries the send glyph, and the footer opens with why it sends and where to turn that off:

one-click card

With the setting on, Start sends the prompt and the new session runs it:

one-click start

With the setting on, in a project that is not a git repository:

one-click no-git card

With the setting on, a refused send shows the server's reason and the retry on the item:

one-click refused

With the setting on, a refused send whose session will not close removes the item and shows one notice that points at the sidebar:

one-click kept

With the setting on, a lost answer says the dashboard could not confirm the prompt reached this session:

one-click unconfirmed

Related Issues

Design: #26380 (rfc-followup-start-sends.md), amended 2026-11-05 in review of 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

@Pearcekieser
Pearcekieser requested review from a team and KiroCrewCaptain as code owners October 29, 2026 20:01
@github-actions github-actions Bot added the readiness: checking Automated validation is still running label Oct 29, 2026
@github-actions

github-actions Bot commented Oct 29, 2026 •

Copy link
Copy Markdown
Contributor

Design Review (Opus 5.5) — ✅ PASS

Design-level review of 0cf21e1e1f6bd3e44e286de55a9cf7c597de3b0c — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

Design-Verdict: PASS

Nothing blocks: the new mode is opt-in, stored only in the browser and never restored from the agent-writable backup, and the default still pre-fills as it does on main.

Notes

  • The two removed footer keys have no readers left outside the locale catalogs. The section says "Compatible." rather than using a Reader: line, but no reader breaks.

[DESIGN-REVIEWED] 0cf21e1

Description read

[DESCRIPTION-READ] c1ffb73738e65990df25d84a73f5727843f1458087dca825b86c0909213a3941

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 29, 2026 •

Copy link
Copy Markdown
Contributor

First Principles Review (Opus 5.5) — ✅ PASS

Premise-level review of 0cf21e1e1f6bd3e44e286de55a9cf7c597de3b0c — 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: PASS

The relabel of "Start in new worktree" to "Start in a separate copy" changes the label for everyone, and the only reason given is a reader nobody names. Nothing blocks the merge.

Notes

  • Item 3 — unjustified move: the only support is "The button says 'worktree', which a first-time reader does not understand". It names no report and no misclick, and no entry under docs/decisions/ records the change (searched for worktree: 0 files). The base RFC does not cover the rename either; only this PR's amendment does. Subtraction: keep the start_in_new_worktree text as it is on main and keep only the new tooltip.

What this change ships

Inventory (8 items) — 7 justified

Intent: let a person who opts in start a follow-up suggestion with one click that sends its prompt, while Start keeps pre-filling by default. This is an ADDITION.

  1. New setting "Send follow-ups with one click", off by default and stored only in the browser — justified
  2. With the setting on, Start and Run send the prompt to the new session before switching to it, and show a send glyph — justified
  3. "Start in new worktree" now reads "Start in a separate copy", for everyone — unjustified move
  4. Each card item shows the first three lines of its prompt, with "Show prompt" for the rest, in both modes — justified
  5. The default footer now says "message box", and the worktree messages now say "copy" — justified
  6. A refused, lost or unconfirmed one-click send shows on its item or in a page notice, and closes the empty session — justified
  7. A one-click send records ↑ history and does not scroll the transcript on screen — justified
  8. The tool description, the model note, the docs, the spec and the RFC amendment describe both modes — justified

[FIRST-PRINCIPLES-REVIEWED] 0cf21e1

Description read

[DESCRIPTION-READ] 6e7d515addc7b04b84dc261b90d251131b44ef1cca2ed80b6e6bd0564b6f75a6

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 29, 2026 •

Copy link
Copy Markdown
Contributor

UX Review (Opus 5.5) — ✅ PASS

UX-level review of 0cf21e1e1f6bd3e44e286de55a9cf7c597de3b0c — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

UX-Verdict: PASS

Nothing blocks. The blind reader read both modes of the primary button correctly. They hesitated before "Run in a separate copy" only until they could read the whole prompt, and the PR's own Show prompt toggle removed that hesitation (shot 06).

Notes

  • Setting name collision: in Settings → Chat → Composer, "Send follow-ups with one click" sits right next to the existing "Quick Send". The reader said: "they sit next to each other and sound alike. I can't tell how they differ." Change the new label to name its scope, for example "Follow-up suggestions: send on click", or add a helper line that sets it apart from Quick Send.
  • The preview's "…" hides whole steps: in shot 01 the cut-off preview gave "no hint that there was a whole extra step hidden". When lines are dropped, make the toggle say how much is hidden, for example "Show full prompt (+2 lines)", in FollowUpCard.tsx show_prompt.
  • Same failure, two looks: a refused send appears inline on the item with a retry (shot 10), but a kept or close-failed send removes the item and shows a page banner (shot 08). The reader "can't tell why the two failures look so different". The banner copy should open with why the item is gone, for example "The new session stayed open, so this suggestion was removed", before it says to look in the sidebar.

[UX-REVIEWED] 0cf21e1

Description read

[DESCRIPTION-READ] cd771964ba227f0ff10d68096c6769e9cd190e9055774e733cba0edab8720f6b

That sha256 names everything this verdict read: the PR description, plus the 18 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 29, 2026 •

Copy link
Copy Markdown
Contributor

GPT 6.1 Review — ✅ no blocking findings

GPT 6.1 completed its review of 0cf21e1e1f6bd3e44e286de55a9cf7c597de3b0c and found no blocking issues.

This comment is updated in place on each push.

Review details

No findings.
[GPT-REVIEWED] 0cf21e1

False positive or not applicable? A repository writer can comment:
/ai-review override gpt 0cf21e1e1f6bd3e44e286de55a9cf7c597de3b0c false-positive: <reason, 20 to 500 characters>

@github-actions

github-actions Bot commented Oct 29, 2026 •

Copy link
Copy Markdown
Contributor

Opus 5.5 Review — ✅ no blocking findings

Reviewed 0cf21e1e1f6bd3e44e286de55a9cf7c597de3b0c — this comment is updated in place on each push.

Review details

No findings.

[OPUS-REVIEWED] 0cf21e1

Verdict parsed from the review's SHA-scoped output markers for commit 0cf21e1e1f6bd3e44e286de55a9cf7c597de3b0c.

False positive or not applicable? A repository writer can comment:
/ai-review override fable 0cf21e1e1f6bd3e44e286de55a9cf7c597de3b0c false-positive: <reason, 20 to 500 characters>

@github-actions github-actions Bot added readiness: action required A blocking check or review needs attention and removed readiness: checking Automated validation is still running labels Oct 29, 2026

@bolichen97 bolichen97 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.

Thanks for this. The goal is clear and the send-before-switch ordering plus the Show prompt toggle are careful work. Requesting changes for these points at 27c3a32:

  1. Product shape. Start now runs agent-authored text on one click, which replaces the pre-fill the base code documents as a deliberate product constraint ("a click can never start an unattended turn"). That is a consent change, so it needs an accepted RFC under docs/request-for-change/ or an explicit maintainer design sign-off before it can merge. The First Principles lane blocks on the same point.

  2. Worktree retry after a refused send cannot succeed. When the send returns false on the worktree route, the scoped session is kept. The retry message tells the user to press Start again, but the reused worktree then hits the in-use check in the worktree create handler (a slot is already scoped to that path) and returns 409 every time. Either delete the just-created slot on a refused send, the way the scoping-failure path does, or retry the send into that same slot, and reword the message and the guide to match.

  3. The owning spec is not updated. docs/system-specs/modules/learn-cron-dashboard.md still says both card actions pre-fill a composer and never send, and that the worktree route activates and then prefills. Please update that paragraph in the same commit: send, clear the card, then switch, plus the not-sent, not-opened and not-focused outcomes and the Show prompt toggle.

Also worth folding in: a transport error after the server has launched the turn makes send() return false, so the card says nothing was sent and invites a second run of the same prompt. Treating an indeterminate result separately from a refusal would avoid that duplicate run.

CI is red on this head (frontend, backend and coverage jobs), so those need a look too.

// launch uses. Resolves false only when nothing was delivered; the card
// item owns that report, so the send posts no page notice of its own.
const delivered = await send(item.prompt, slotKey, undefined, true, true)
if (!delivered) throw new Error(fail.notSent)

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.

A refused send on the worktree route leaves a retry that always fails.

When send returns false here, the new slot stays scoped to the worktree. The not-sent message asks the user to press Start again, the create call reuses the worktree, and the handler's in-use check finds that slot scoped to the same path and refuses with 409. The prompt can never be sent from the card after the first refusal.

Suggestion: on a refused send, delete the slot just created (as the scoping-failure path does), or retry the send into this same slot, then make the message and the guide say what actually happens.

Separately, a transport error after the server launched the turn also returns false here, so a retry can run the prompt twice. Consider telling "refused" apart from "unknown".

@Pearcekieser
Pearcekieser force-pushed the pearcekieser/feat-followup-start-sends branch from 27c3a32 to e90a02e Compare October 30, 2026 01:03
@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 30, 2026
@Pearcekieser

Pearcekieser commented Oct 30, 2026 •

Copy link
Copy Markdown
Contributor Author
  • A refused send blocks the worktree retry / indeterminate delivery offers a duplicate-run retry (span=0c0b3c42651b)

fixed in e90a02e: openFollowupSession now closes the empty session (deleteSlot) only after a send that never reached the server, so a retry reuses the worktree instead of hitting worktree_in_use.
A lost receipt (transport-error) is reported to the caller as indeterminate through a new send() callback; the start then continues as delivered, keeps the session, marks the bubble unconfirmed and shows a check-the-session notice instead of a retry.

@Pearcekieser

Pearcekieser commented Oct 30, 2026 •

Copy link
Copy Markdown
Contributor Author
  • The navigation origin is captured after worktree creation (span=0c0b3c42651b)

fixed in e90a02e: the start-time slot and the store's user-move counter (activeSlotMoves) are captured in followupOrigin, before the createWorktree await, and the switch is skipped if either changed.
Tests cover a switch during the worktree create and a leave-and-return during the send.

@Pearcekieser

Pearcekieser commented Oct 30, 2026 •

Copy link
Copy Markdown
Contributor Author
  • Refused send on the worktree route leaves the session scoped, so every retry is refused as in use (span=01dfac6df927)

fixed in e90a02e: a send that never reached the server now closes the session it created, and the retry message says the worktree is reused. A test asserts deleteChatSlot runs for the new session.

@Pearcekieser

Pearcekieser commented Oct 30, 2026 •

Copy link
Copy Markdown
Contributor Author
  • Owning spec still says Start never sends (span=e58ee26efa2a)

fixed in e90a02e: docs/system-specs/modules/learn-cron-dashboard.md now describes send, clear the card, then switch, the not-sent, unconfirmed, not-opened and not-focused outcomes, and the Show prompt toggle.

@Pearcekieser

Pearcekieser commented Oct 30, 2026 •

Copy link
Copy Markdown
Contributor Author
  • Product-shape change without accepted RFC (span=6e00ebc99189)

needs-a-decision: Start sending the prompt reverses the documented pre-fill constraint. This was requested by the issue author; the PR body now names the reversal. A maintainer needs to record the decision, either an accepted RFC or /ai-review override first-principles <head> product-shape: <decision>.

@Pearcekieser

Pearcekieser commented Oct 30, 2026 •

Copy link
Copy Markdown
Contributor Author
  • The main button does not say what it now does (span=d8d092c4fa31)

needs-a-decision: relabelling Start (for example "Run in new session (worktree)") and adding a files-untouched line beside it is a UI-placement change the issue did not settle. It waits on the same maintainer decision as the product-shape item.

@Pearcekieser

Copy link
Copy Markdown
Contributor Author

question (blocking): product shape. Agreed this reverses a recorded consent rule. Pearce asked for Start to run the prompt, and I have asked him to get a maintainer decision or open an RFC. Would a design sign-off here from you be enough, or do you want an RFC under docs/request-for-change/ first?

issue: worktree retry after a refused send. Fixed in e90a02e. A send that never reached the server now closes the session it created, so pressing Start again reuses the worktree. The retry message and the guide say so.

issue: owning spec. Fixed in e90a02e. The learn-cron-dashboard.md paragraph now covers send, clear the card, then switch, the not-sent, unconfirmed, not-opened and not-focused outcomes, and Show prompt.

suggestion: lost receipt. Done in e90a02e. A transport error is now unconfirmed, not refused: the card clears, the session is kept, its first message shows the pending-delivery mark, and a notice says to check that session, so no second run is offered.

note: the red CI on 27c3a32 came from tests this PR does not touch; they pass after the rebase onto current main.

@github-actions github-actions Bot added readiness: action required A blocking check or review needs attention and removed readiness: checking Automated validation is still running labels Oct 30, 2026
@Pearcekieser

Pearcekieser commented Oct 30, 2026 •

Copy link
Copy Markdown
Contributor Author
  • Preserve pending Start state across card remounts (span=0c0b3c42651b)

fixed in f6353bb: in-flight starts are tracked in useFollowupStarts, which stays mounted across session switches, keyed by origin slot, card ts and prompt, so a click on the remounted card does not start the prompt a second time.
A start that fails after the user left the origin, even if they came back, also posts a page notice, so the error is not lost with the unmounted card.

@Pearcekieser
Pearcekieser force-pushed the pearcekieser/feat-followup-start-sends branch from e90a02e to f6353bb Compare October 30, 2026 01:31
@Pearcekieser

Pearcekieser commented Oct 30, 2026 •

Copy link
Copy Markdown
Contributor Author
  • The primary Start control is unreadable and now acts at once (span=b5eed277fae5)

needs-a-decision: renaming "Start in new worktree" to a label that names the outcome, and adding a files-untouched line next to it, is a UI-placement change the issue did not settle. It waits on the same maintainer decision as the product-shape item; the lane passed this head's predecessor e90a02e with the same label.

@Pearcekieser

Pearcekieser commented Oct 30, 2026 •

Copy link
Copy Markdown
Contributor Author
  • Product-shape change without accepted RFC (span=6e00ebc99189)

needs-a-decision: unchanged since the last head. Start sending the prompt reverses the documented pre-fill rule and needs a maintainer to record the decision, an accepted RFC or /ai-review override first-principles f6353bbaf9ae0415a64f653c5c2b60f450a69134 product-shape: <decision>.

@bolichen97 bolichen97 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.

Re-reviewed at f6353bb against our earlier request at 27c3a32.

Fixed:

  1. Worktree retry after a refused send. A refused send now deletes the just-created slot before reporting not-sent. The worktree create handler builds its in-use list from live slot projects, so the retry no longer hits the 409 "a session is already open in it" refusal. The new tests assert the delete on refusal and no delete on an unconfirmed send.
  2. Owning spec. docs/system-specs/modules/learn-cron-dashboard.md now says Add to this session pre-fills and Start sends, and lists the refused, unconfirmed, not-opened and not-focused outcomes plus the Show prompt toggle.

Also addressed: a lost receipt is now treated as unconfirmed rather than refused, so the card no longer invites a second run of a prompt that may already be running.

Minor, non-blocking: the delete on a refused send swallows its own failure. If that delete fails, the card still says to press Start again and the worktree retry returns 409 as before. Consider a different message in that case, or leave as is if you judge it rare enough.

Still open: the product-shape question. Start now runs agent-authored text on one click instead of pre-filling the composer, and that needs an accepted RFC or an explicit maintainer design sign-off. That decision is still pending with the maintainers, so this is a comment rather than an approval. The GPT lane finding on the duplicate-start guard resetting on route navigation also still stands on this head.

@Pearcekieser
Pearcekieser force-pushed the pearcekieser/feat-followup-start-sends branch from f6353bb to aef533c Compare October 30, 2026 02:17
@github-actions github-actions Bot added the merge conflict Branch has merge conflicts with its base — author must resolve before merge label Nov 6, 2026
@Pearcekieser
Pearcekieser force-pushed the pearcekieser/feat-followup-start-sends branch from 76c05e1 to e578b40 Compare November 7, 2026 02:12
@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 7, 2026
@Pearcekieser

Copy link
Copy Markdown
Contributor Author
  • The one-click Run button is understood, but readers won't press it (span=a9a6ffebe6f2)

fixed in e578b40: with one-click send on, the card footer now opens with the reason and the way back: "Sends right away because Send follow-ups with one click is on. Turn it off in Settings." The second sentence is a SettingsLink to /settings/chat/composer?highlight=chat.send-follow-ups-with-one-click, which opens the toggle highlighted. Both sentences are in English and the 11 translations, and the default pre-fill footer is unchanged. FollowUpCard.test.tsx checks the text and the link target in one-click mode, and that default mode shows neither. after-send-1-card.png and after-send-nogit-card.png show it.

@Pearcekieser

Copy link
Copy Markdown
Contributor Author

@bolichen97 the Run-mode footer change from your 06:12Z review is in on e578b40.

note: With one-click send on, the footer now opens with "Sends right away because Send follow-ups with one click is on. Turn it off in Settings." The second sentence is a SettingsLink to Settings, Chat, Composer with the toggle highlighted (/settings/chat/composer?highlight=chat.send-follow-ups-with-one-click). The setting name in the sentence is read from the settings label key, so the two cannot drift. The rest of the footer is unchanged, and the default pre-fill footer has no notice and no link. Both sentences are translated in the 11 catalogs. FollowUpCard.test.tsx checks the text and the link target in one-click mode, and that default mode shows neither. after-send-1-card.png shows it.

note: The branch is rebased onto current main again. Main changed deleteSlot's close-refused handling (closeConfirmations, watchCloseOutcome), so I resolved that conflict by hand: the caller-reported close keeps skipping the sidebar notice and the new close watch, and every other close is unchanged. ChatSlice.uncodedClose.test.tsx covers both.

@github-actions github-actions Bot added merge conflict Branch has merge conflicts with its base — author must resolve before merge readiness: passed Eligible automated validation passed for the current revision 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 7, 2026
@bolichen97

Copy link
Copy Markdown
Contributor

The branch now conflicts with main. Could you rebase it onto current main, keeping one or two commits for PR Hygiene?

In the same push, please take the open point from my last review. With one-click send on, the Run-mode footer still does not say that this behavior comes from a setting, or where to turn it off. At 5715df6, footer_start_sends_add_prefills reads "Run in new chat sends the prompt as the first message of a new chat. Add to this session puts the prompt in your message box and sends nothing." Please start it with the reason and the way back, for example "Sends right away (Send follow-ups with one click is on; change it in Settings)", ideally linking the setting name to Settings → Chat → Composer.

I'll approve with auto once the rebased head has that and is green.

@Pearcekieser

Copy link
Copy Markdown
Contributor Author

note (non-blocking): rebased onto main as one commit, and each Run-mode footer string now opens with the reason and the way back.

Head 1900001 sits on main eb0ecb9. In one-click mode the three footer strings (footer_worktree_start_sends, footer_start_sends_add_prefills and footer_start_sends_add_prefills_project_directory) now start with "Sends right away because Send follow-ups with one click is on. Turn it off in Settings → Chat → Composer." The setting name links to /settings/chat/composer?highlight=chat.send-follow-ups-with-one-click. The card renders each string through Trans, so the reason, the link and the description are one string in all 12 catalogs, and the Settings path is built from the nav labels.

At 5715df6 the card drew that lead from two separate keys placed before the string you quoted, so the string alone did not say it. Those keys (footer_one_click_on and footer_one_click_turn_off) are removed. The one-click screenshots in the description were recaptured from this head.

@Pearcekieser

Copy link
Copy Markdown
Contributor Author
  • The "separate copy" Start/Run button scares people off (span=fdf87c3de7f4)

not changed: the button labels and both footer_worktree_* strings on 47ecf5a are identical to f18601d, where this lane passed at 04:45Z and said the blind reader understood every primary control. The only change between the two heads is the App budget line, which this head now takes from main (#33665). The copy wording is the one bolichen97 asked for on Nov 6: the footer says the copy is a fresh copy of this project in a new folder next to it, and that current files and this session stay as they are. A third round of footer copy across 12 catalogs would answer a verdict that flipped on unchanged text, so the wording stays as the maintainer set it.

resolution: narrowed -- the worktree copy wording stays as the maintainer set it; further footer copy changes are out of scope for this PR

@Pearcekieser

Copy link
Copy Markdown
Contributor Author
  • Second quiet close of an empty session (span=cbf9ab4ee155)

fixed in 0bc0f98: the refused-send cleanup in useChatPageSlotSwitches.ts now calls main's deleteSlot({ key, discard: true }) from #32350. deleteSlotCallerReported and its module Set are gone, so chatSlice.ts, its public-surface and ownership pins and ChatSlice.uncodedClose.test.tsx are identical to main. Only a discard that closed the session itself offers the retry; a refusal or a 404 keeps the session and points at it in the sidebar. The tests assert discardChatSlot, with a 404 case added.

@Pearcekieser

Copy link
Copy Markdown
Contributor Author
  • The new worktree label still scares people off its main action (span=29f9d063fd4c)

not changed: "Start in a separate copy", "Run in a separate copy" and both footer_worktree_* strings are the same on d6cde3b as on f18601d, where this lane passed and said the blind reader understood every primary control. The wording is the one bolichen97 asked for on Nov 6. The suggested "Delete it from the sidebar anytime" is not true today: closing the session leaves the worktree folder, and the only removal is the Dev Fleet app or git, so saying it would promise a removal the dashboard does not offer.

@Pearcekieser

Copy link
Copy Markdown
Contributor Author
  • Readers won't use the one-click Run button (span=987d323cd42a)

not changed: the Run labels, the send glyph and the one-click footer on 69457a0 are identical to 9970f77, where this lane passed at 08:34Z on Nov 9. The rebase between the two changed only the generated UI index and one test assertion. The footer opens with the reason and the way back, as bolichen97 asked on Nov 8, and the setting is off by default, so only a person who turned it on sees Run. Asking the maintainer for a UX override on this head.

@Pearcekieser

Copy link
Copy Markdown
Contributor Author

@bolichen97 question (non-blocking): could you record a UX override for 69457a0?

The UX lane has now flipped three times on copy that did not change. It passed f18601d, blocked 47ecf5a and d6cde3b, passed 9970f77, and blocked 69457a0. The labels, footers and send glyph are byte-identical across all five heads, and the rebases between them touched only generated files, the bundle budget and test assertions. Every other lane passes on 69457a0: Design, First Principles, GPT 6.1 and Opus 5.5.

The footer opens with "Sends right away because Send follow-ups with one click is on. Turn it off in Settings → Chat → Composer.", as you asked on Nov 8, and the setting is off by default. If you would rather change the wording first, tell me which line and I will push it.

The command is /ai-review override ux 69457a0b898a7d4581f584aa4aeb9cd595d8fc6c false-positive: <reason>.

@Pearcekieser

Copy link
Copy Markdown
Contributor Author
  • Run sends text hidden behind "Show prompt" (span=d991b748ee03)

not changed: the three-line preview is Decision 3 of the accepted RFC (docs/request-for-change/rfc-followup-start-sends.md), and its Open questions section records that showing the prompt by default "was settled in review: the first three lines are shown, the rest behind Show prompt". The RFC lists the case this finding names as a known risk: a prompt whose real instruction sits after line three. The preview code is the same on 8ec1bd4 as on 9970f77, where this lane passed. Showing the full prompt in one-click mode would reverse a maintainer decision, so it is put to bolichen97 rather than changed here.

@Pearcekieser

Pearcekieser commented Nov 10, 2026 •

Copy link
Copy Markdown
Contributor Author

Update (08:40Z): UX passed on 0cf21e1, so no override or change is needed. Every lane is green on that head (Design, First Principles, GPT 6.1, Opus 5.5 and UX), and all 97 checks pass.

If you approve, please leave auto-merge to me. Main has moved 36 commits since this head's base, so I will rebase onto it first and arm auto-merge once that head is green.

Earlier question, kept for the record and no longer needed:

@bolichen97 question (non-blocking): the rebase onto main moved the head to 8ec1bd4, so the UX override I asked for on 69457a0 no longer applies. UX now blocks for a different reason: in one-click mode, Run sends a prompt whose fourth line onward is behind Show prompt. Design, First Principles, GPT 6.1 and Opus 5.5 all pass on this head.

The three-line preview is Decision 3 of the RFC, settled in your review, so I have not changed it. Which do you want?

  1. Keep the RFC as is and record the override: /ai-review override ux 8ec1bd43aa548af6f8955c489c40d11397f86243 false-positive: <reason>.
  2. Show the full prompt, with no Show prompt toggle, only when one-click send is on, as UX suggests. I would push that with a line in the RFC amendment, and pre-fill mode keeps the three-line preview.

@KiroCrewCaptain

Copy link
Copy Markdown
Collaborator

The design is signed off: the RFC rfc-followup-start-sends.md was accepted on 2026-10-31, and the opt-in amendment follows the maintainer's review on this PR. The PR now waits for a maintainer approval. The open change request on 76c05e1 also needs a re-review. At this head, the one-click footer opens with the reason and the setting name, which is what that review asked for.

What the review checked on 0cf21e1:

  • The default pre-fill path matches main. Only an explicit send mode sends.
  • The send goes out before the switch and is addressed to the new session by key.
  • A refused send, a kept session, a failed close and a lost or late receipt each give the result the RFC failure table describes, and each has its own test.
  • A remount or a double click sends once.
  • The backend edits are text only.

It found no blocking issue. The failure table is turned into tests one row at a time, and the consent fence is tested from the attacker's side as well.

For the security review:

  • The opt-in is stored only in the browser, under mc-followup-start-sends. Only the stored value "1" turns it on. It is not in DURABLE_PREF_KEYS and has no configKey, so neither the agent-writable settings backup nor an agent change card can turn it on. Tests pin both.
  • The prompt is sent with optionText, so a prompt that starts with a slash command does not run a UI command.
  • The prompt is rendered as plain text only.
  • With the setting on, the card shows only the first three lines of the prompt, as the RFC says. The rest is behind Show prompt.

Optional:

  • src/kiro_crew/dashboard/chat_api/followup_card.py:123: the warning the model sees still names the old "Start in new worktree" button. Consider the "Start in a separate copy" wording used in session_directive_apply_slot.py.
  • website/src/components/FollowUpCard.tsx:82: the preview counts blank lines. A prompt that starts with three empty lines shows only "…". Consider skipping leading blank lines, with a test.
  • website/src/test/ChatPageW3Coverage.test.tsx grows to 1,673 lines, past the 1,600-line split target. Consider moving the follow-up describe block into its own test file.

@bolichen97 bolichen97 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.

Approving on 0cf21e1. All 97 checks are green, including UX Review and PR Readiness.

The last point from my review at 76c05e1 is fixed. Each one-click footer (footer_worktree_start_sends, footer_start_sends_add_prefills, ..._project_directory) now opens with "Sends right away because Send follow-ups with one click is on. Turn it off in Settings → Chat → Composer.", and the setting's own name is a SettingsLink to that control. The path is built from the tabs' own labels, so it matches the nav, and the spec says the same.

Also in this head, compared against 76c05e1: a refused send now discards the empty session through deleteSlot({ discard: true }) and offers the retry only when that discard really closed it. A 404 or a refusal counts as kept, so a retry never meets a session that comes back holding the copy. The sidebar "could not close" notice stays quiet there, because the kept-session notice already says what to do.

The design is the one agreed in review and in rfc-followup-start-sends.md. Pre-fill stays the default for both buttons. One-click send is opt-in under the browser-local mc-followup-start-sends, outside the host backup. No regex in product code, no new restrictions.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

readiness: passed Eligible automated validation passed for the current revision

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants