fix(clients): align placeholders with conversation state

[serge-the-hedge]

Input placeholders on desktop/web seemed incoherent with what's going on, and the copy wasn't great too.

Turned out they were following "connected / disconnected" state instead of more obvious "new / existing conversation" state that was already used on mobile. I was annoyed more than I probably should've been, so here we go.

	Before		After
Surface	New thread	Existing thread	New thread	Existing thread
Mobile	“Describe a coding task in {project}”	“Ask the repo agent, or run a command…”	"Describe the task, tag @files, use $skills or /commands"	“Ask for changes, add context, or attach images”
Web	"Ask for follow-up changes or attach images" (because state is usually "disconnected")	“Ask anything, @tag files/folders, $use skills, or / for commands” (because state is usually "connected")
Desktop
Font preview

What Changed

Now there is canonical copy for new and continuing conversation states that lives in the shared client runtime. Web, desktop, and mobile, all invoke the same copy.

• New conversations show: “Describe the task, tag @files, use $skills or /commands”
• Existing conversations show: “Ask for changes, add context, or attach images”

The new copy:

a. Provides a more meaningful tooltip for native
b. Stylistically more consistent than the old desktop one

Web/desktop selects between the two from the rendered conversation, including optimistic messages; mobile explicitly selects the appropriate copy for its new-task and existing-thread composers.

Why

Web/desktop previously selected its ordinary placeholder from session phase. A new draft has no session and was treated as disconnected, so it showed follow-up guidance. Once a session existed, continued conversations showed the basic starter instructions – the opposite of the intended behavior.

UI Changes

The composer placeholder copy changes on web, desktop, and mobile.

Verification

• vp run --filter @t3tools/client-runtime typecheck
• vp run --filter @t3tools/web typecheck
• vp run --filter @t3tools/mobile typecheck
• Targeted lint for every changed TypeScript file
• Full repository test run: 7,405 tests passed; two unrelated root-runner suites failed because a .cjs script was collected without tests and the Ghostty ABI suite could not import its vendored wasm?inline asset

Checklist

• This PR is small and focused
• I explained what changed and why
• I included before/after screenshots for the UI changes
• A video is not applicable because no motion or interaction timing changed

Implemented with GPT-5.6 Sol via Codex in the T3 Code harness.

Note

Standardize composer placeholder text based on conversation state

• Adds a conversationComposerPlaceholder function in composer.ts that maps a ComposerConversationKind ("new" or "existing") to a shared guidance string.
• Adds resolveComposerConversationKind in ChatView.logic.ts to derive the kind from message count, loading state, and latestUserMessageAt.
• Web ChatComposer and mobile composers (NewTaskDraftScreen, ThreadDetailScreen) now use these shared helpers instead of hardcoded or interpolated placeholder strings.
• Behavioral Change: the web composer no longer shows phase- or state-specific placeholder variants; all clients now show one of two standardized strings.

^{Macroscope summarized c24b879.}

---

Note

Low Risk
Copy and lightweight UI classification only; no auth, data, or send-path changes.

Overview
Composer placeholders now follow new vs existing conversation instead of web/desktop session connected/disconnected, with one shared copy path for web, mobile, and settings previews.

A new @t3tools/client-runtime/composer export defines conversationComposerPlaceholder for 'new' and 'existing' strings. Web derives kind via resolveComposerConversationKind (timeline message count, thread detail loading, and shell latestUserMessageAt so loading threads still get follow-up copy) and passes conversationKind into ChatComposer, replacing the old phase-based default placeholder. Mobile new-task and thread composers and the settings font preview call the same helper with explicit 'new' / 'existing'.

^{Reviewed by Cursor Bugbot for commit c24b879. Bugbot is set up for automated code reviews on this repo. Configure here.}

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Repository UI

Review profile: CHILL

Plan: Pro Plus

Run ID: 8d02314d-9c0c-4056-9c4c-a1c71dea202f

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[cursor]

Cursor Bugbot has reviewed your changes using high effort and found 1 potential issue.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

Want fixes drafted automatically? Bugbot Autofix can create code changes for findings. A team admin can enable Autofix in the Cursor dashboard.

^{Reviewed by Cursor Bugbot for commit c60bd52. Configure here.}

[macroscopeapp]

Approvability

Verdict: Approved c24b879

This PR consolidates composer placeholder text into a shared utility and adds context-aware selection (new vs existing conversation). Changes are limited to UI copy in input fields with no business logic impact. Unit test coverage included.

^{You can customize Macroscope's approvability policy. Learn more.}

Dismissing prior approval to re-evaluate c900bcc