## Context
We previously introduced a behaviour for the explorer home tab's chat
form to run a SQL Query if the input is detected to be a SQL query.
Adjusting it to shift that behaviour into a secondary action instead
<img width="740" height="210" alt="image"
src="https://github.com/user-attachments/assets/d4b1fec1-f38c-426f-8108-b50ead1a61ca"
/>
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* SQL statements entered in Explorer can be run directly with a
dedicated “Run SQL” action.
* Assistant forms support context-specific submit icons, labels,
tooltips, and accessibility text.
* **Bug Fixes**
* Improved SQL detection for multi-statement queries.
* Prevented mixed SQL and conversational text from being treated as
executable SQL.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
Five layout fixes across Explorer chat, notebooks, and the sidebar.
### Chat
- **Conversation fade overlapped the scrollbar.** The top and bottom
gradients are positioned against the conversation's padding box, which
includes the scroll container's scrollbar gutter, so `inset-x-0` painted
them over the scrollbar. They now stop at the conversation's content
gutter, which `Conversation` owns for both the content and the fades.
- **Composer background bled past the input's radius.** The form paints
the surface behind the textarea but had no radius of its own, so its
square corners showed outside the `rounded-lg` input. It now shares the
radius.
- **Message parts used two different widths.** Wide parts come down to
`max-w-3xl` so every part shares a column, matching `AssistantQueryCell`
and `AssistantNotebookPreview`. `isWide` / `isWideMessagePart` stay in
place with both widths equal, so a part can diverge again later without
rebuilding the mechanism.
### Notebooks
- **Cell controls sat at the container edge.** Each cell centred itself
at its own max width while the grip and add-cell button stayed at the
far left of the full-width row, leaving a large gap. `SortableSection`
takes a `sectionWidth` and carries its control gutter twice — once as
the controls, once as padding on the other side — so the section stays
centred with its controls immediately beside it. Cell widths are
unchanged (prose `48rem`, query `72rem`); set them equal and the two
cell types' controls line up on their own.
The controls stay in flow rather than floating in an outside gutter, so
on a viewport narrower than the cap the row just fills the space instead
of clipping the controls into the padding.
### Sidebar
- **Search icon didn't line up with the menu row icons.** The row box
already sits flush with the search input's box, so rows moved from
`pl-3` to `pl-2` to put their icons on the same 8px offset the search
icon uses. Spacing between the input and the list now matches the 12px
side padding.
### Testing
`pnpm --filter studio run typecheck`, Prettier, and 378 tests across
`Explorer`, `ProjectHome`, `AIAssistantPanel`, and `ExplorerLayout`
pass. ESLint warning counts are unchanged from master.
These were reasoned from layout rather than checked in a browser, so
they're worth a look on a preview before merge.
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **UI Improvements**
* Updated Explorer layouts with flexible, configurable widths for
notebook and query sections.
* Refined navigation spacing and padding across Explorer views.
* Centered and standardized AI Assistant preview, query, and message
content widths.
* Improved chat form styling with rounded corners.
* Adjusted conversation spacing and fade overlays to avoid overlapping
the scrollbar.
* Preserved full-width behavior where appropriate while keeping controls
aligned.
* **Tests**
* Updated layout tests to reflect revised width and alignment behavior.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
## I have read the
[CONTRIBUTING.md](https://github.com/supabase/supabase/blob/master/CONTRIBUTING.md)
file.
YES
## What kind of change does this PR introduce?
UI refinements for Explorer query surfaces.
## What is the current behavior?
- The assistant chat textarea uses a tighter radius than the Run SQL /
Create a notebook cards on Explorer home.
- Chart results sit unevenly in the results pane because axis gutters
stack on card padding, and long Y labels can clip.
- Selecting SQL in the editor changes the primary Run button to Run
selected, which is easy to trigger by accident.
- The Prettify SQL icon in notebook query cells uses Lucide's default
size, so it doesn't match other toolbar actions.
## What is the new behavior?
- Assistant chat form uses `rounded-lg` so it matches the home action
cards everywhere the form is used.
- Query result charts collapse unused axis space, add a little padding
when labels are on, and size the Y axis from formatted ticks so longer
labels fit.
- Run is a default split button that always executes the full query. Run
selected is a secondary menu item, disabled until SQL is selected.
- Notebook cell Prettify SQL icons use `size={16}` and `strokeWidth={2}`
like the rest of the Explorer toolbar.
## Additional context
Cmd+Enter in the editor still runs the current selection when there is
one.
## Test plan
- [ ] Open Explorer home and confirm the assistant chat radius matches
the Run SQL and Create a notebook cards.
- [ ] Run a query, switch to chart view, and check spacing with labels
off and on, including large Y values.
- [ ] With no selection, click Run and confirm the full query runs. Open
the split menu and confirm Run selected is disabled.
- [ ] Select SQL, click Run, and confirm the full query still runs. Use
Run selected from the menu to run only the selection.
- [ ] In a notebook query cell, confirm Prettify SQL matches the size
and stroke of nearby toolbar icons.
Made with [Cursor](https://cursor.com)
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
- **New Features**
- Added a split Run control in the query editor, with separate actions
for running all content or only selected text.
- Added support for customizing chart X-axis display settings.
- Improved chart Y-axis sizing, scaling, and tick formatting for clearer
results.
- **Bug Fixes**
- The “Run selected” action is unavailable when no text is selected.
- **Style**
- Updated toolbar icon sizing and added rounded corners to the assistant
chat input.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
## I have read the
[CONTRIBUTING.md](https://github.com/supabase/supabase/blob/master/CONTRIBUTING.md)
file.
YES
## What kind of change does this PR introduce?
Feature / chore: hide assistant model selection in the UI and default
chats to GPT-5.6 Luna.
## What is the current behavior?
The assistant composer exposes a model picker. Paid orgs default to
`gpt-5.3-codex`; everyone else defaults to `gpt-5.4-nano`.
## What is the new behavior?
- The model picker is hidden in the assistant composer and Explorer
home.
- Chats default to `gpt-5.6-luna` with `reasoningEffort: medium`.
- Model selection plumbing is kept (registry, entitlements, `setModel`,
generate-v4 request body) so a requested model can still be honored when
provided.
- Other completion endpoints still use `gpt-5.4-nano`.
## Additional context
Model selector UI can be re-enabled by passing `selectedModel` /
`onSelectModel` to `AssistantChatForm`.
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Added support for the GPT-5.6 Luna model with medium reasoning
capability.
* Made GPT-5.6 Luna the default assistant model.
* **Improvements**
* Simplified assistant chat by removing model selection from the primary
chat experience.
* Updated model fallback behavior to use the standard assistant model.
* Chat forms can now optionally display model selection when configured.
* **Tests**
* Updated model coverage and assistant chat tests for the new defaults
and behavior.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Co-authored-by: Cursor <cursoragent@cursor.com>
This PR removes all `paths` in `tsconfig.json` for all apps and
packages. They were added previosly because some of the components had a
`_Shadcn` suffix because of an ongoing migration. How that the migration
is done, the paths can be removed.
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Refactor**
* Standardized shared UI component, utility, and icon imports across
design-system examples and application screens.
* Simplified shared component access and project configuration.
* Added shared access to anchor-link helpers and animation styles.
* **Compatibility**
* Updated component exports and imports without changing existing
behavior.
* No changes to user-facing workflows, screens, or functionality.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
## Context
More groundwork for the Explorer - initializing the home page
Note that nothing here is functional, all just visual still
<img width="1387" height="960" alt="image"
src="https://github.com/user-attachments/assets/d4967578-edbd-476f-8150-d9d5e9d66666"
/>
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Added a new Explorer landing page with an assistant chat form.
* Added quick actions for creating notebooks and SQL work.
* Added notebook and chat template cards for faster project exploration.
* **Improvements**
* Explorer content now fills the available page height.
* Assistant send button styling now reflects whether submission is
available.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
## I have read the
[CONTRIBUTING.md](https://github.com/supabase/supabase/blob/master/CONTRIBUTING.md)
file.
YES
## What kind of change does this PR introduce?
Feature — final PR (9/9) of the SQL editor logs-source stack.
**Base branch:** `charislam/sql-editor-inline-ai-clickhouse-dialect` (PR
8). Nothing here is user-visible: entry points stay behind
`sqlEditorLogsSource` + `otelLegacyLogs`, and flag rollout happens after
the whole stack merges.
## What is the current behavior?
- The Assistant has no idea a SQL editor snippet targets the logs
backend. Ask it about a logs snippet and it answers in Postgres, because
the attached query is fenced as ` ```sql ` and nothing tells the model
otherwise.
- Because the `sql` fence is what `MessageMarkdown` treats as runnable
Postgres, an attached ClickHouse query is rendered with a
Run-against-Postgres affordance and branded with `untrustedSql`.
- "Debug with Assistant" on a failed logs query produces a dialect-less
prompt, so both the in-app assistant and the copyable version get
debugged as Postgres.
- A report referencing a `log_sql` snippet runs its ClickHouse SQL
against the user's Postgres database and surfaces the resulting error.
## What is the new behavior?
**Assistant panel.** The "Current Query" chip records which backend the
attached query targets. That reaches the model two ways: each attachment
is fenced with its own dialect (` ```clickhouse ` vs ` ```sql `), and a
`containsLogsSnippets` flag rides on the user message as AI SDK
`metadata`. The server reads the flag off the conversation and prepends
the ClickHouse dialect rules plus the logs schema reference as a
non-cached context message.
Two design points worth calling out in review:
- The flag lives on the **message**, not the request body, so Retry and
the tool-approval continuation reproduce the context a message was
originally asked in — neither of those passes a per-call body.
- It's derived from **what's actually attached**, so detaching the chip
drops the claim rather than leaving the two able to disagree.
The `clickhouse` fence also keeps a logs query out of
`MessageMarkdown`'s `sql` branch, so it's no longer offered as runnable
Postgres or branded with `untrustedSql` — a boundary this stack's
distinct brands exist to prevent crossing.
**Debug flow.** `buildDebugChatArgs` attaches its query with a source
for the same reason, and names the dialect in the prompt text so the
copyable version stands on its own outside the app.
**Reports.** A report only stores a snippet id, so whether it queries
the logs backend is only knowable once the content loads. `ReportBlock`
guards on the fetched type and renders a `LogsSnippetReportBlock`
placeholder instead of executing. Double-guarded: no `sql` for a logs
snippet (so it's out of the query key and `queryFn` short-circuits even
on an explicit `refetch`) and `enabled` excludes it.
**Incidental cleanups.** `buildAssistantContextMessages` extracted out
of `generate-assistant-response`; a schema-access sentinel that was
duplicated as a string literal across two files (and compared against)
replaced with one exported constant; `SqlSnippet` deduplicated to a
single declaration; `resolveSnippetSource` / `isLogsSource` shared
instead of re-implemented per surface.
**Tests.** 4 new/extended suites. Notable cases pinned: a message with
no metadata must validate (`safeValidateUIMessages` applies
`metadataSchema` to *every* message, so a required schema would 400
every existing conversation); only *user* messages count, so a model
reply can't talk the server into a different dialect; a mixed-attachment
message is flagged without overclaiming a single source; and
`ReportBlock` registers no pg-meta mock for the logs cases, so an
unhandled request failing the test *is* the assertion that logs SQL
never reaches Postgres.
Verified: `pnpm typecheck`, `lint:ratchet` (no regression), Prettier,
and the full Studio suite (459 files / 4969 tests).
## Additional context
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
- **New Features**
- Added support for recognizing log snippets in reports, with clear
guidance to open them in the SQL editor or remove them.
- AI Assistant now understands log snippets and provides
ClickHouse-specific context, formatting, and troubleshooting guidance.
- Snippets retain their source information when shared with the AI
Assistant.
- **Bug Fixes**
- Prevented unsupported log snippets from being executed as regular
database queries.
- Improved source detection when opening snippets directly from links.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
## Context
Part of some minor improvements to the AI Assistant - this one's about
adding some keyboard shortcuts
## Changes involved
- Added keyboard shortcut for "New chat"
<img width="212" height="96" alt="image"
src="https://github.com/user-attachments/assets/e9c3bd63-adbc-4b05-8c52-1baed67e365a"
/>
- Also added a small animation for the "How can I assist you?" text for
visual indication when moving between chats that might not have a
conversation yet
- Added keyboard shortcut for "Permission settings"
<img width="236" height="86" alt="image"
src="https://github.com/user-attachments/assets/337da009-5979-4910-9292-73cc4d7f7cce"
/>
- Show keyboard shortcut for "Close Assistant"
<img width="165" height="88" alt="image"
src="https://github.com/user-attachments/assets/b45a4f5a-8e12-45f1-8fdb-a18c0deedb02"
/>
- Fix `ExpandingTextArea` height calculation logic issue
- If you open and close the Assistant panel a number of times, the
height of the input field isn't consistent, so this fixes that
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
## Summary by CodeRabbit
- **New Features**
- Added keyboard shortcuts for starting a new AI Assistant chat and
opening permission settings.
- Header actions now display shortcut hints and support keyboard access.
- **Improvements**
- Enhanced accessibility with labels for chat edit controls (save,
cancel, edit, delete).
- Chat onboarding now remounts when switching active chats.
- Improved chat popover alignment.
- Escape now blurs the message input; textarea resizing is more reliable
during content/layout changes.
- **Bug Fixes**
- Updated onboarding loading behavior based on the lints loading state.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
## Problem
Our `<Button>` component breaks the default `button` contract by
redefining the `type` prop to set its variant (`primary`, `default`,
etc) instead of the button type (`submit`, `button`, etc).
This is confusing and forces to write more code when using it with
shadcn components that expect/inject the standard button props.
## Solution
- rename the `type` prop to `variant`
- rename the `htmlType` prop to `type`
- propagate the changes where necessary
- format code
## How to test
As this is just prop renaming, if it builds it's ok
---------
Co-authored-by: Ivan Vasilov <vasilov.ivan@gmail.com>
When Assistant requests confirmation to run a query or deploy an edge
function if the user doesn't skip or run and instead sends a follow-up
message it errors out. This allows follow-up messages and treats them as
"skips" which means adjusting confirmation message state as part of the
follow-up. This also uses toModelOutput to cleanse data based on
permissions.
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
## Release Notes
* **New Features**
* Enhanced tool approval workflow: pending approvals are now
automatically resolved as denied when submitting new messages
* Improved chat input state management with better handling of approval
states
* Customizable loading messages for tool operations
* **Bug Fixes**
* Fixed chat input availability during pending tool approval states
* Improved tool execution feedback during approval workflows
<!-- review_stack_entry_start -->
[](https://app.coderabbit.ai/change-stack/supabase/supabase/pull/46052?utm_source=github_walkthrough&utm_medium=github&utm_campaign=change_stack)
<!-- review_stack_entry_end -->
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
- Most changes are related to either types or `useRef` usages (it now
requires an initial value).
- also updated `vaul` to its latest version and haven't noticed any
change ([design-system
demo](https://design-system-git-react-19-supabase.vercel.app/design-system/docs/components/drawer))
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Upgraded workspace to React 19.
* **Bug Fixes**
* Improved null-safety and ref handling across editors, UI components,
shortcuts, and markdown/image rendering to reduce runtime errors.
* Safer event/timeout/interval cleanup and more robust command/context
handling.
* **Chores**
* Bumped vaul dependency versions.
* **Documentation**
* Type and TypeScript accuracy improvements for clearer developer
feedback.
<!-- review_stack_entry_start -->
[](https://app.coderabbit.ai/change-stack/supabase/supabase/pull/45886)
<!-- review_stack_entry_end -->
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
We're exploring support for newer models like
[gpt-5.4-nano](https://openai.com/index/introducing-gpt-5-4-mini-and-nano/)
in Assistant. This model doesn't support the `'minimal'` reasoning
effort level we use for gpt-5-mini which leads to vague errors.
<img width="595" height="263" alt="CleanShot 2026-03-18 at 17 13 05@2x"
src="https://github.com/user-attachments/assets/cf7c2370-322d-4a8a-be55-23e680db0aa0"
/>
Also, we've [previously
discussed](https://supabase.slack.com/archives/C0161K73J1J/p1771544464850199?thread_ts=1771493920.775699&cid=C0161K73J1J)
that reasoning adds unnecessary latency to otherwise simple AI
completion endpoints like `title-v2`. We want more control of reasoning
level independent of model/endpoint.
This PR aims to solve both problems by:
- making reasoning effort configurable on a per-request basis
- adding compile-time guardrails to prevent selecting an incompatible
reasoning level for models
- adding a `DEFAULT_COMPLETION_MODEL` with minimal reasoning that we can
update with newer models that support disabling reasoning (independent
of Assistant chat model reasoning)
Other improvements to our model config logic:
- Fixes bug in `onboarding/design.ts` and `assistant.eval.ts` where
`providerOptions` was being dropped
- `getModel()` now returns a bundled `modelParams` object (spread into
AI SDK calls) so `providerOptions` can't be accidentally omitted (this
[has happened
before](https://supabase.slack.com/archives/C0161K73J1J/p1771518443534309?thread_ts=1771493920.775699&cid=C0161K73J1J))
- Introduces an `ASSISTANT_MODELS` registry as a single source of truth
for assistant model config, eliminating hardcoded model IDs across the
codebase
- Aligns free/pro model conditional logic with `assistant.advance_model`
entitlement naming conventions instead of the `isLimited` pattern
- Adds `console.error` logging of Assistant stream errors so we can
interpret reasoning effort compatibility errors in the future (instead
of just opaque "Sorry, I'm having trouble responding right now" card)
- Removes unnecessary type casts and generally making the model config
logic stricter
- Removes pre-existing dead code: `anthropic` provider variant in
`GetModelParams` / `PROVIDERS` registry that was never implemented in
`getModel()`
Now if you try to select an unsupported reasoning level you get a type
error:
<img width="1306" height="320" alt="CleanShot 2026-03-20 at 14 37 24@2x"
src="https://github.com/user-attachments/assets/a6ac234b-5ea5-4d81-8e01-ac4be34a0800"
/>
And if for some reason an invalid reasoning level slips through, you now
get a server-side error surfacing the issue:
<img width="1268" height="204" alt="CleanShot 2026-03-20 at 14 58 14@2x"
src="https://github.com/user-attachments/assets/aadc1b7a-9495-475f-9741-39979bd27cd7"
/>
I've tested gpt-5 and gpt-5-mini are still working on the staging
preview and verified the models were selected properly in Braintrust
logs. Both models are available on my Pro test account, and my Free test
account shows the Pro upgrade CTA.
Closes AI-446
Closes AI-551
* add delete chat message functionality
* Implement message editing functionality in AIAssistant
* add functionality to delete messages in AI Assistant chat in assistant state
* remove duplicated `text-xs` class
* add a 'group' class for improved styling and refining action button visibility on hover
* Refine action button visibility in AIAssistantPanel to show only for user messages on hover, improving user experience.
* remove `"` from the message to edit
* Enhance editing message display in AIAssistantPanel by updating styling for improved visibility and user experience.
* Refactor AIAssistantPanel state management by simplifying original message content initialization and clarifying comments for better code readability.
* clean up action button rendering in AIAssistantPanel message component
* adjust the chat message layout (space)
* simplify the editing mode message
* show the edit button even when user is already in the edit mode
* Update AssistantChatForm styling for editing mode, positioning the editing message above the input field without changing the layout
* use `ButtonTooltip` for edit button to show tooltip
* simplify the condiiton to show the edit buton
* add the delete action button
* Refactor action button rendering in AIAssistantPanel to improve clarity and maintainability. Ensure buttons are only shown for user messages on hover, and simplify the conditional rendering logic.
* Remove commented-out code for cleaner implementation.
* improve unused onDeleteAfter prop and consolidating delete logic into a single `deleteMessageFromHere` function for improved clarity
* abort streaming when a user deletes a message
* delete unnecessary `originalMessageContent` state variable
* delete `confirmEdit` and move its logic into `sendMessageToAssistant`
* add `strokeWidth={1.5}` to `Pencil` icon
* add `opacity-50` to any messages that are after the message being edited, and it cancels the edit when those messages are clicked
* add a pulse animation to the user’s message being edited
* add more space between mesages and `AssistantChatForm`
* add framer-motion animation to `Editing message` element
* update strokeWidth of Trash2 icon in Message component for better visibility
* refactor Message component to make onDelete prop required
* add `DeleteMessageConfirmModal` for message deletion confirmation in Message component
* refactor message deletion logic in AIAssistant component to streamline editing process
* cancel editing when a user clicks the edit button again or the edit button in subsequent messages
* Add success toast notification for message deletion
* add Pencil icon to AIAssistantPanel component and fixed duplicated tailwind class
* Update AIAssistantPanel to use the Pencil icon
* Refactor AIAssistant to set editing message content from text parts
* Remove unused `reload` prop from AIAssistant component
* `setIsSticky(false)` when a user deletes all the messages or clear the chat to remove the `ArrowDown` icon scroll down button
* refactor: remove deleteMessagesBefore function from AiAssistantState
* refactor: simplify AssistantChatForm layout by removing unnecessary `z-index`
* enable the input field when a user wants to edit the message while the response is streaming
* Smol fixes
* remove conosle log
* Add state for resubmitting in AI Assistant panel
* Remove classes that hide the focus styles + fix focusing when editing message to ensure cursor is at the end of the text
---------
Co-authored-by: Joshen Lim <joshenlimek@gmail.com>
* style: update `ExpandingTextArea` focus styles
* revert the changes on `index.tsx`
* add `focus:border-control` to text-area component
* remove `overflow-hidden` from `form` to show box-shadow
---------
Co-authored-by: Terry Sutton <saltcod@gmail.com>
* Implement message editing functionality in AIAssistant
* add functionality to delete messages in AI Assistant chat in assistant state
* remove duplicated `text-xs` class
* add a 'group' class for improved styling and refining action button visibility on hover
* Refine action button visibility in AIAssistantPanel to show only for user messages on hover, improving user experience.
* remove `"` from the message to edit
* Enhance editing message display in AIAssistantPanel by updating styling for improved visibility and user experience.
* Refactor AIAssistantPanel state management by simplifying original message content initialization and clarifying comments for better code readability.
* clean up action button rendering in AIAssistantPanel message component
* adjust the chat message layout (space)
* simplify the editing mode message
* show the edit button even when user is already in the edit mode
* Update AssistantChatForm styling for editing mode, positioning the editing message above the input field without changing the layout
* use `ButtonTooltip` for edit button to show tooltip
* simplify the condiiton to show the edit buton
* delete unnecessary `originalMessageContent` state variable
* delete `confirmEdit` and move its logic into `sendMessageToAssistant`
* add `strokeWidth={1.5}` to `Pencil` icon
* add `opacity-50` to any messages that are after the message being edited, and it cancels the edit when those messages are clicked
* add a pulse animation to the user’s message being edited
* add more space between mesages and `AssistantChatForm`
* add framer-motion animation to `Editing message` element
* cancel editing when a user clicks the edit button again or the edit button in subsequent messages
* add Pencil icon to AIAssistantPanel component and fixed duplicated tailwind class
* Update AIAssistantPanel to use the Pencil icon
* Refactor AIAssistant to set editing message content from text parts
* Minor changes
* nit
---------
Co-authored-by: Joshen Lim <joshenlimek@gmail.com>
* get `stop()` from useChat and use it
* Implement abort signal in `generate-v4` API
* Enhance `onStop` functionality in AIAssistant to store partial message when aborted
* Update dependencies in AIAssistant to include `updateMessage` in effect hook (lint warning fix)
* change onStop prop to be required in AssistantChatForm
* add comment for saving partial AI responses
* remove redundant text size class
* Make onStop prop optional in AssistantChatForm and conditionally render stop button
* Add loading spinner to stop button in AssistantChatForm
* simplify opacity handling for submit button
* Add tooltips
* Improve UI error handler for assistant panel
* Small tweak
* Filter out any tool calls that are streaming in generate-v4, happens when aborting the assistant while a tool is being called
---------
Co-authored-by: Joshen Lim <joshenlimek@gmail.com>
Joshen Lim
