## Problem
- observability report modals still uses `formik` and we want to remove
it in favour of `react-hook-form` to keep only one form library
## Solution
- Migrate to `react-hook-form`
## 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?
Docs update to `humans.txt` to add `Jordan McQueen`
## What is the current behavior?
`Jordan McQueen` does not yet exist in `humans.txt`
## What is the new behavior?
`Jordan McQueen` exists in `humans.txt`
## 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?
docs update
## What is the current behavior?
Uses the phrase: "For e.g." instead of "For instance,"
## What is the new behavior?
Substituted out the abbreviation "e.g." for the word "instance"
---------
Co-authored-by: Chris Chinchilla <chris.ward@supabase.io>
## What kind of change does this PR introduce?
UI improvements. Resolves DEPR-419.
## What is the current behavior?
- `ColumnList` splits type information across separate `Data Type` and
`Format` columns, which makes scanning less efficient
- The list only surfaces a limited subset of column metadata up front,
so key structural information like primary keys, foreign keys,
uniqueness, and identity is not easy to scan
- Constraint and nullability affordances are not aligned with the visual
language already used in the Table Editor and Schema Visualiser
## What is the new behavior?
- `ColumnList` now shows a leading type affordance icon, while keeping
the exact Postgres type visible in the `Type` column
- The standalone `Data Type` column is removed, and `format` is now the
primary textual type shown in the table
- The `Constraints` column now surfaces explicit icon-plus-label tokens
for:
- `Primary`
- `Foreign key`
- `Unique`
- `Identity`
- `Nullable` / `Non-nullable`
- Constraint tokens use a purpose-built segmented style based on
`ComputeBadge`, including a green primary variant for primary keys
- Nullability now matches Schema Visualiser semantics exactly: outlined
diamond for nullable, filled diamond for non-nullable
- Existing search, descriptions, row actions, and permission gating are
preserved
---------
Co-authored-by: Joshen Lim <joshenlimek@gmail.com>
## Context
Refactors the Table Editor's `ColumnEditor` to use the latest UI
components, and fix some UI oddities along the way
## Bug fixes
- Fix header text vertical alignment
- Before:
<img width="325" height="59" alt="image"
src="https://github.com/user-attachments/assets/e4bc07d4-2630-4a86-a87c-4bbbf94e2f52"
/>
- After:
<img width="351" height="74" alt="image"
src="https://github.com/user-attachments/assets/d0a0a246-59b6-4d19-8674-8cc5eb33772c"
/>
- Fix closing a toast would close the panel as well
- Can verify by creating a new column, then hitting save without
entering anything. Will trigger some error toasts and closing them will
close the panel too
- Fix tooltips on "is nullable" and "is unique" showing up irregardless
if "is primary key" is toggled on or off
### Changes
- (shouldFetch): `!org.billing_partner` - prevents the unnecessary API
call to fetch the customer profile for partner orgs.
- (early return guard): `org.billing_partner` - returns null (hides the
banner) when the org has a billing partner.
## What kind of change does this PR introduce?
UI improvements
## What is the current behavior?
- The database tables list and columns list use inconsistent page shells
and table primitives
- The child columns page has weaker information hierarchy and row
actions than the parent tables page
- Responsive column priority on the tables list does not reflect the
most important data on smaller breakpoints
- Table actions and counts are harder to scan than they should be
## What is the new behavior?
- Both pages now use `PageLayout` with matching large-width content
containers
- `ColumnList` now uses the latest `ui` Table primitives instead of the
legacy cleaned-up-later table
- Both pages now show totals in a table footer
- `ColumnList` now uses a tiny filter input, case-insensitive filtering,
inline descriptions under the name, and a primary `Edit` button with
overflow actions
- `TableList` now has improved responsive column priority:
- smallest breakpoint keeps `Rows`
- `Columns` appears from `sm`
- `Size` appears from `lg`
- `Realtime Enabled` appears from `2xl`
- `TableList` now uses `View columns` as the CTA, removes the ambiguous
icon from that CTA, restores the entity icon from `sm` upwards only, and
tightens the name column on the smallest breakpoint only
- Boolean icon columns are right-aligned consistently, with the same
Realtime icon tones applied to both `Realtime Enabled` and `Nullable`
- The columns detail page now uses breadcrumbs for navigation back to
Tables instead of an inline back button
| Before | After |
| --- | --- |
| <img width="1728" height="997" alt="Tables Database Mallet Toolshed
Supabase-0E0E3DE0-4EA1-407F-88D4-B85664D26D8E"
src="https://github.com/user-attachments/assets/3a2e265c-394e-432c-8c29-12317b60fda8"
/> | <img width="1728" height="997" alt="Tables Database Mallet Toolshed
Supabase-C8FC339C-E9DA-4ADB-8458-C7EFF55F2AEC"
src="https://github.com/user-attachments/assets/50c83a3f-a70c-4d09-a8c3-1eeaed68b68b"
/> |
| <img width="1728" height="997" alt="Tables Database Mallet Toolshed
Supabase-FE9196A0-BEAF-4BA5-8A2C-06F934A62C38"
src="https://github.com/user-attachments/assets/707a564a-e764-45ac-8470-8532e22d39bc"
/> | <img width="1728" height="997" alt="Tables Database Mallet Toolshed
Supabase-36E93C1E-7943-4C98-8119-CAF48E2FE5BA"
src="https://github.com/user-attachments/assets/4cba5791-a4d7-4f43-aea0-8277b2ec5d28"
/> |
---------
Co-authored-by: Joshen Lim <joshenlimek@gmail.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?
Investigation by claude, validated!
- Before (Formik): The old form used `<Form
initialValues={INITIAL_VALUES} validationSchema={...}>` Formik does not
unregister hidden fields — all fields from initialValues stay in form
state with their initial values, so hidden required fields still pass
validation because they retain their default values (e.g., the OTP
expiry/length numbers from the config).
- After (react-hook-form): The new form uses useForm({ shouldUnregister:
true }). This explicitly removes fields from form state when their
components unmount. When Twilio Verify is selected, the three hidden
fields are unmounted, their values become undefined, and yup's
unconditional .required() fails silently.
this bug was introduced today by PR #44095, the
Formik-to-react-hook-form migration.
## Problem
Users cannot create nor edit foreign keys when editing a table or a
column from the schema visualiser
## Solution
The foreign key edition component is always reading the current table
from the URL parameters. However, in the context of the schema
visualiser, the parameter does not exist.
Allows to override it from the side panel editor context
## How to test
- Create two tables
- Go to the schema visualiser and edit one of the table
- Click the _Add foreign key relation_ and add a foreign key to the
other table
- Edit one of the foreign key column
- You should be able to edit the relation
## I have read the
[CONTRIBUTING.md](https://github.com/supabase/supabase/blob/master/CONTRIBUTING.md)
file.
YES/NO
## What kind of change does this PR introduce?
fixes: #37454
The context menu on the FK row is a lie, its just exposing the cell
underneath. Ideally this should have its own context menu, right now we
only support copy operations.
---------
Co-authored-by: Joshen Lim <joshenlimek@gmail.com>
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
## 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?
When a project's redirect URL list exceeds the 2KB client-side limit,
the preflight length check in onConfirmDeleteUrl prevents users from
removing URLs, showing "Too many redirect URLs, please remove some or
try to use wildcards" on delete operations.
This creates a deadlock where users can neither add nor remove URLs.
The check validates the payload after removing the selected URLs, but if
the remaining list is still over 2KB, it blocks the request. Since
removing URLs can only reduce payload size (never increase it), this
check is unnecessary on the delete path.
## 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?
- Index mismatch on ADD then DELETE/EDIT:
formatGridDataWithOperationValues was searching the original rows array
for DELETE_ROW and EDIT_CELL_CONTENT operations, then using those
indices on the modified formattedRows array (which had been shifted by
ADD_ROW's unshift). Both now search formattedRows directly.
- Cross-table operation leaking: The entire operation queue was passed
to formatGridDataWithOperationValues without filtering by the current
table, causing pending ADD_ROW and DELETE_ROW operations from one table
to appear in other tables. Operations are now filtered by tableId before
rendering.
Francesco Sansalvadore