DB migration compat / Back-compat — Current branch migrations with ${{ needs.check-migrations-changed.outputs.base_branch }} branch code (push) Has been cancelled
DB migration compat / Forward-compat — Current branch code with ${{ needs.check-migrations-changed.outputs.base_branch }} branch migrations (push) Has been cancelled
DB migration compat / Back-compat — Current branch migrations with ${{ needs.check-migrations-changed.outputs.base_branch }} branch code (push) Has been cancelled
DB migration compat / Forward-compat — Current branch code with ${{ needs.check-migrations-changed.outputs.base_branch }} branch migrations (push) Has been cancelled
## Summary
**Stacked on #1468** (`docs/hexclave-rename-plan` — the plan doc). Diff
vs that base = the actual PR 1 code.
This is **PR 1 of the Hexclave rebrand: the invisible compatibility
layer**. Everything is additive. Old SDKs, old wire identifiers, and old
env var names keep working unchanged. The backend dual-accepts and
dual-emits; new SDK code emits `x-hexclave-*` headers and the
`hexclave_` Bearer prefix; cookies dual-write; env vars dual-read across
every category. **No user-visible rebranding lands here** — that's PR 2.
See [`RENAME-TO-HEXCLAVE.md`](./RENAME-TO-HEXCLAVE.md) → *"PR 1
implementation guide"* for the full per-work-area spec, file pointers,
and chosen approach.
## What's implemented (all 14 PR-1 work-areas)
- **SDK export aliases** — `Hexclave*` aliases for the user-facing
`Stack*` exports added in `packages/template`; codegen propagates them
to `@stackframe/{js,stack,react,tanstack-start}`. React-only aliases
correctly excluded from `@stackframe/js`. (`e60550a2`)
- **JWT issuer dual-accept** — `decodeAccessToken` accepts both
`api.stack-auth.com` and `api.hexclave.com` issuers. Signing unchanged.
(`fc781def`)
- **Request-header dual-accept** — backend + dashboard proxies normalize
`x-hexclave-*` → `x-stack-*` at the existing empty proxy hook (so
`smart-request.tsx` and every route schema keep working unchanged); CORS
allowlists extended via a derive-once helper. (`2a056eac`)
- **MCP `ask_hexclave`** — registered alongside `ask_stack_auth` via a
shared helper; `ask_stack_auth` behavior byte-identical. (`30ffd604`)
- **Dev-tool** — DOM ids + header emit switched.
`window.HexclaveDevTool` exposed alongside `window.StackDevTool`.
(`32131ea7`)
- **The big consolidated commit** (`7fed864a`):
- **Env vars** — central `getEnvVariable` prefix-transform (HEXCLAVE
first, STACK fallback); dashboard + template client env files dual-read;
`turbo.json` globalEnv; `NEXT_PUBLIC_STACK_PORT_PREFIX` renamed outright
across ~82 files including docker.
- **Cookies** — dual-write/dual-read auth (`stack-access`/`-refresh-*`
and custom-domain variants), OAuth-state
(`stack-oauth-{inner,outer}-*`), and low-risk cookies (`stack-is-https`,
`stack-last-seen-changelog-version`). Bypass sites patched (backend
OAuth callback, dashboard remote-dev auth route, impersonation snippets,
snapshot serializer).
- **Bearer prefix** — SDK token parser accepts both `stackauth_` and
`hexclave_`; emits `hexclave_`. Discovery correction: this is purely
SDK-internal — the backend never parses it.
- **Response headers** — backend dual-emits
`x-hexclave-{request-id,actual-status,known-error}`; SDKs dual-read (new
first, stack fallback).
- **SDK request-header emit switch** —
`client/server/admin-interface.ts` + dashboard `api-headers.ts` +
`internal-project-headers.ts` + `feedback-form.tsx` switched to
`x-hexclave-*`. Plus `stack_response_mode` query param.
- **Storage keys** — dev-tool / cli-auth / oauth-button / docs keys
renamed (straight); `stack:session-replay:v1` dual-read so in-progress
recordings survive SDK upgrades; `stack_mfa_attempt_code` dual-read.
- **Query params** — cross-domain params dual-emit/dual-accept via
shared helpers; backend `oauth/authorize` accepts
`hexclave_response_mode` and `stack_response_mode`; `stack-init-id`
renamed.
- **`Symbol.for`** — app-internals symbol gets a parallel
`Symbol.for("Hexclave--app-internals")` getter on each attach site (no
read-site churn — old symbol still attached). 3 file-private symbols
renamed outright.
- **Config discovery** — prefer `hexclave.config.ts`, fall back to
`stack.config.ts` at every discovery site (CLI / dashboard / backend /
local-emulator); `init` writes the new filename; CLI credentials path
migrates.
- **Internal renames** — `StackAssertionError`,
`StackClient/Server/AdminInterface` renamed outright (no alias, per the
"internal-only → rename" rule). ~264 files touched.
- **Review-pass fixes** (`21217fbe`) — three real bugs found by parallel
review agents and fixed:
- `snapshot-serializer.ts` was interpolating the whole
`keyedCookieNamePrefixes` array (`${arr}`) — adding a second prefix
would have corrupted **every** OAuth-cookie snapshot, not just new ones.
- **Docker port-prefix producer/consumer mismatch** —
`entrypoint.sh`/`run-emulator.sh`/cloud-init `user-data` were still
producing `NEXT_PUBLIC_STACK_PORT_PREFIX` while the dashboard sentinel +
consumers had been renamed; silent self-host regression (custom port
prefix would be ignored).
- **Missing `hexclave-oauth-inner-*` dual-write** in the OAuth authorize
route — callback's fallback masked it but the dual-write was specified
by the plan.
- Plus: `mcp.test.ts` tool-list assertions updated to include
`ask_hexclave`; two dashboard header-emit sites switched to
`x-hexclave-*` for consistency.
- **E2E snapshot serializer follow-up** (`4b16cc5d`) —
`x-hexclave-request-id` added to the hidden-headers list (mirroring
`x-stack-request-id` treatment), and 2 sample inline snapshots
regenerated in `projects.test.ts` to include the new dual-emitted
headers.
## Verification
- **`pnpm typecheck`** — clean (the fresh-worktree `@/.source` / Prisma
codegen gap in `stack-docs` is pre-existing and unrelated).
- **`pnpm lint`** — 29/29 packages green.
- **`pnpm exec turbo run build --filter=./packages/*`** — 13/13 packages
build (including `@stackframe/stack-cli` once the dashboard standalone
is present).
- **Live E2E** against a running backend on `cl/hexclave-pr1`:
- `pnpm test run
apps/e2e/tests/backend/endpoints/api/v1/internal/mcp.test.ts` — **6/6
pass** (verifies the new `ask_hexclave` tool — the hand-written inline
snapshot matched actual MCP server output).
- `pnpm test run
apps/e2e/tests/backend/endpoints/api/v1/internal/projects.test.ts` —
**11/11 pass** (verifies wire dual-accept + dual-emit end-to-end; the
snapshot serializer fix was found and applied during this check).
A four-agent parallel **review pass** also audited the full diff for
logic/runtime bugs across the work-areas (wire headers + JWT, cookies +
bearer + symbols, env vars, query params + config + MCP + aliases). All
in-slice review verdicts were ✓ except the three bugs listed above,
which are now fixed.
## Known follow-ups (out of scope for this PR)
- **E2E snapshots across the rest of the suite** — backend now
dual-emits `x-hexclave-{known-error,actual-status}` alongside
`x-stack-*`, which legitimately appears in inline snapshots throughout
`apps/e2e`. Two were regenerated here as a sample; the rest should regen
with `vitest -u` in CI.
- **Docker shell env vars beyond `PORT_PREFIX`** — `entrypoint.sh` still
reads `STACK_*` env vars directly (the JS-side `getEnvVariable`
transform doesn't help the shell). JS consumers dual-read so it works in
practice; full shell-level dual-read is a deeper self-host follow-up.
- **`@stackframe/stack-cli` build ordering** — pre-existing; needs
`build:rde-standalone` first. Not affected by this PR.
## Test plan
- [ ] CI runs full e2e suite (with `vitest -u` to absorb dual-emit
snapshot deltas, then committed back)
- [ ] Spot-check: an old SDK build (emitting only `x-stack-*`) still
authenticates against the new backend
- [ ] Spot-check: a new SDK (emitting `x-hexclave-*` / `Bearer
hexclave_*`) still authenticates against an old backend during deploy
ordering
- [ ] Manual: `npx @stackframe/stack-cli@latest init` (new onboarding
entrypoint) generates `hexclave.config.ts`
- [ ] Manual: existing `stack.config.ts`-only project still resolves (no
migration required)
---------
Co-authored-by: bilal <bilal@stack-auth.com>
DB migration compat / Back-compat — Current branch migrations with ${{ needs.check-migrations-changed.outputs.base_branch }} branch code (push) Has been cancelled
DB migration compat / Forward-compat — Current branch code with ${{ needs.check-migrations-changed.outputs.base_branch }} branch migrations (push) Has been cancelled
### Context
We didn't have an easy place for a user to see their domain statistics
and track their sent emails, either overall or by draft. Additionally,
there was scope creep with the sidebar, where we were supporting more
pages. Our emails landing page was also rather confusing, especially
toggling/ working with different email server types. So, we decide to
add a "sent" page, to track email logs and email statistics, as well as
let users temporarily override their sending limits if need be.
Additionally, a user may want to see a particular email in more detail:
what stage is it in? How did it proceed through time? How can I pause
the sending of this email or change the scheduled time or edit the code?
We allow for that to happen.
### Summary of Changes
#### New Pages
1. **Sent Page:** A Domain Reputation card lets you track how many of
your sent emails were bounced or marked as spam as well as how much
capacity you have left. We also provide a temporary override, where you
can use up to 4 times your capacity for a limited period of time.
Additionally, we provide an email log that lets you see the recently
sent emails. You can also toggle this view from a "list all emails" to
"group by template/draft" which shows stats for each template/draft id
(i.e a bar showing how many emails were sent, are pending, were marked
as spam, were bounced etc, and the total number of emails sent with that
template or draft). Clicking on an email in the list all view takes you
to the "email-viewer" endpoint for that email (see below). Clicking on a
template/draft in the group by view takes you to a page where you can
see the statistics for that template/draft in more detail (the "send"
stage view for that template/draft, as referenced below).
2. **Settings Page:** This is a new page we created because the old
"emails" landing page wasn't doing its job. This page is to track all
the email settings. Currently, we put in 2 sections. A "theme settings"
card where users can see their active theme and click on a button to be
navigated to the themes page. This is necessary as we remove themes from
the sidebar. The other section is a card for email server and domain
configuration - you can change your server type and adjust the settings
or send a test email. It's cleaner and less noisy.
3. **Drafts Page**: There are a lot of changes here. On the landing
page, we actually separate out the drafts into "active drafts" and
"draft history" because drafts are meant to be fire-and-forget, not
reusable. We also add the functionality to create a draft from a
template. This was tricky to manage because templates rely on template
variables which sent to the backend along with the code and injected
during render time. We deal with this by having AI rewrite the template
source code to remove any references to template variables and to make
the draft standalone. The drafts page has been separated into a
stepper-controlled multi stage process:
draft->recipients->schedule->sent. Sent is a read only view that shows
you the statistics of the emails sent using that draft, as mentioned
earlier. You can also see the sent view of a historical draft. You can
also bulk pause/cancel any unsent emails from the sent view of the
drafts.
4. **Sidebar Updates**: The email sidebar now doesn't show "themes" or
"emails" (the old landing page), but it does show "settings" and "sent",
and the default landing page for emails is "sent".
5. **Email Viewer**: When you click on an individual email, you get
navigated here. This has a timeline showing the progress of the email on
the right, and some optional info for the user that's toggleable on the
right bottom, while having either a preview of the email if it's sent or
a way to edit it. You can also change the scheduledAt date of an email
if it hasn't already been sent.
#### Bug Fixes
1. **Search in `TeamMemberSearchTable`**: This was broken. Every time
you tried to enter or remove a character, it would trigger skeleton
loading that overlapped the search bar too, preventing you from
adding/removing more. This was caused because the `useUser` hook
eventually ended up calling a `use` hook, which throws a promise that
triggers a suspense. This, coupled with the fact that the implementation
of `TeamMemberSearchTable` involved a prop-drilling/ dependency
inversion approach to passing down its toolbar to a base table
component, meant the suspense would cover the toolbar too and couldn't
be scoped to just the table. A refactor has gotten rid of the need for
those base components while fixing tables in `payments/customers`,
`teams/team_id`, and `payments/transactions` on top of the existing use
in email drafts recipients stage. We also dedupped some code.
2. **Stale draft fetches on draft landing page**: `useEmailDrafts` uses
an asyncCache to cache the fetched drafts. It is used on the drafts
landing page to render the drafts. When a draft is sent, its `sentAt` is
marked versus when it is still active, it is marked as null. The cache
was stale and so navigating to the landing page after firing off a draft
would errorneously represent that draft as still active and indeed, even
allow you to edit it and fire it again. This violated the principle of
drafts being fire and forget. This has been dealt with by adding
functionality to refresh the draft cache upon firing off a draft.
#### Other Changes
1. We bumped up the base time for the exponential send attempt retry
backoff in `email-queue-step` to 20 seconds. The previous base was two
seconds, and this effectively just made it wait until the next iteration
of the `email-queue-step` cron job or at most an iteration that wasn't
too far away. When an outage with our provider happens, it may take a
while for it to be resolved, so a longer backoff is justified
2. We transitioned the themes page and the templates page to using the
new components, though deeper UI refactors for them were out of scope
for this ticket.
3. We implement a "temporarily increase capacity" button, that bumps up
the throughput/ capacity limit fourfold for a user for a given period of
time. It works like this:
> Clicking the button sets a boost expiredat time.
> When this time is set and still valid, the capacity rate is multiplied
by 4.
> When the button is clicked, trigger a loading spinner until the route
finishes processing.
> When the timer runs out, we reset the button back to its original
state.
> We dont need to wrap the onclick with runAsyncWithAlert because the
component does that already.
4. We add a new default theme: a colorful theme with a lavender base.
This was mainly done so we could have three times in a theme showcase in
the settings page.
### UI Demos
**Sent Page Demo:**
https://github.com/user-attachments/assets/19294a90-bb65-4f00-9a97-111f6c08287f
**Drafts Page Demo**
https://github.com/user-attachments/assets/847609ef-d699-470c-a699-297bb9e17f04
**Settings Page Demo**
https://github.com/user-attachments/assets/190a3829-036a-4f57-89c0-a873bef5a7ce
**Email Viewer Page Demo**
https://github.com/user-attachments/assets/3bc50159-4acb-4865-a4dd-830c84ee4235
---------
Co-authored-by: Konstantin Wohlwend <n2d4xc@gmail.com>
### Summary of Changes
Just bumped up polling, removed unnecessary wait checks in tests that
don't need them. Minor changes, not an exhaustive list of flaky test
fixes
Note that importing a function into a file B that was exported from a
test file A causes vitest to see all the tests in test file A as being
under file B. This messes up CI and makes it harder to track failing
tests.
### Context
Some of our users' emails were getting stuck in sending. The long delays
in processing the retries caused a vercel function timeout.
### Summary of Changes
We refactor the low level email sending functions to remove the retry
logic there. We kick it up to the email queue step. Additionally, we
flag emails to be retried when they encounter issues but leave it for a
future iteration to actually perform the retry. We perform an
exponential backoff with a random component to decide when they have to
be retried. We also make some small adjustments to the queuing function
to not queue skipped emails.
When an email fails to send during the sending function, we check to see
if it is a retryable error or not. Some errors are transient and trying
again may succeed while others indicate deeper issues. If it is
retryable, and the max number of retry attempts hasn't been reached, we
set `nextSendRetryAt` to a time determined by an exponential backoff
calculation function. When the queuing function looks for emails to
queue, it doesn't just pick up the `SCHEDULED`. emails whose
`scheduledAt` time <= `NOW()`, but also those emails whose
`nextSendRetryAt` time <= `NOW()`. What this means in practice is that
one iteration of the `email-queue-step` will mark emails as retryable
while another iteration will perform the retry. This should be cleaner
and prevent long delays in the `email-queue-step` process due to
retries. This also makes it easier to scale up the number of retries if
need be.
### Context
We noticed some errors pop up on sentry related to email rendering.
These errors seem to have been triggered by the same issue, and could be
categorized as follows:
1. Sanity test mismatch, even when the errors from freestyle and vercel
sandbox were broadly similar. This occurred due to stack traces
differing in different execution environments.
2. Rendering errors from freestyle and vercel sandbox caused by the
theme not being imported/ empty theme component.
Upon investigation, this occurred because hitting save on the email
themes page with an invalid theme (ex: deleting the `export` keyword, or
renaming the `EmailTheme` component) still triggers `bundleAndExecute`
with the invalid themes. This will obviously fail and cause the errors
to be logged, however there is no cause for concern here because the
error is returned and the save is denied because an error is returned.
It's more of a matter of noisy error logs and too strict sanity test
comparisons.
Beyond that, `js-execution` is a little opaque and hard to understand,
and this can mask errors in logic.
We also noticed a new issue: manually throwing an error in the email
theme code editor, and then trying to save was actually successful. This
was because the version of `react-email/components` we were using had
faulty error handling, and fell back to client side rendering, masking
the error. This wasn't caught by our `try-catch` safeguards because it
was a render time issue that was masked. More specifically, this was
what `react-email` was doing: `Switched to client rendering because the
server rendering errored`.
### Summary of Changes
We loosen the sanity test comparison between engine execution results in
case of errors. We then refactor the `js-execution` and
`email-rendering` files to read better, and to only `captureError` when
a service is down, but not for runtime errors in the user submitted
code.
To deal with the other bug, we bumped `react-email/components` to the
latest version. However, doing so exposed a gap between real `freestyle`
and our `freestyle-mock`: with the mock, the errors that were now raised
were treated as uncaught exceptions, crashing the mock server.
Consequently, we switched to using `node` over `bun`.
We also expanded test coverage to account for different error paths.
Co-authored-by: Konstantin Wohlwend <n2d4xc@gmail.com>
### Summary of Changes
We had an issue with a test being exceedingly flaky. This was a result
of not enough time having passed between the email being put up for
rendering/ being evaluated for skipped status and it being checked.
Polling reduces our dependency on arbitrary timeout periods. This is in
line with how we poll the inbox in other tests (see
`waitForMessagesWithSubject` in `helpers.ts`). It doesn't solve
flakiness because flakiness is a result of the indeterminism of what
we're testing.
- restricted users
- onboarding app
- waitlist app
- fixed an exception when setting primary email
- automatically update the JWT token on the client when the user object
changes
