mirror of
https://github.com/windmill-labs/windmill.git
synced 2026-08-18 16:02:10 +00:00
fix/hook-tests-worktree-root
14432 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 hugocasa
|
67efefbf53 |
test(agents): make the tree-root rows follow the checkout kind
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
 Ruben Fiszel
|
8efede55d6 |
chore(main): release 1.792.0 (#10745)
* chore(main): release 1.792.0 * Apply automatic changes --------- Co-authored-by: rubenfiszel <275584+rubenfiszel@users.noreply.github.com>v1.792.0 |
||
 Guilhem
|
d40a446868 |
style: use subtle Button for raw app preview toolbar actions (#10747)
* style: use subtle Button for raw app preview toolbar actions * fix: expose inspector toggle pressed state via aria-pressed |
||
|
Ruben Fiszel
|
ef4dc46d4b |
fix(cli): keep script settings on push and repair the up-to-date check (#10741)
* fix(cli): keep script retention, debounce and cache settings on push Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * test(cli): surface the create response when the fixture fails Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * test(cli): drop debounce settings the CI build refuses to accept Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * repair the script push up-to-date comparison (#10743) * test: settle the backlog before the capped audit-export drain (#10737) * test: settle the backlog before the capped audit-export drain * chore: update ee-repo-ref to bd4de74eb37b32a2b6c7c69f6dedac031ef8436b This commit updates the EE repository reference after PR #730 was merged in windmill-ee-private. Previous ee-repo-ref: b5a5f9114df26088cfe976d91f10e55ba8bfcaa6 New ee-repo-ref: bd4de74eb37b32a2b6c7c69f6dedac031ef8436b Automated by sync-ee-ref workflow. --------- Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com> * fix(cli): repair the script push up-to-date comparison Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * test(cli): drain dependency jobs and pin a non-1 priority skip Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * test(cli): describe the priority fixture without the old comparison Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor(cli): read cache_ignore_s3_path off the typed response Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(cli): stop redeploying bunnative scripts on every push Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com> |
||
|
Guilhem
|
7b17e358b3 |
feat(frontend): record the outcome of every AI chat tool call (#10746)
* feat(frontend): record the outcome of every AI chat tool call The `ai_chat`/`tool` counter fired before execution, so nothing recorded whether a tool call succeeded, and the three paths that refuse a call before it runs recorded nothing at all. Log once per call on whichever path ends it, keyed `<tool_name>:<status>` over ok, error, declined, rejected and blocked_plan_mode. Per-tool totals now need `split_part(key, ':', 1)` downstream; rows keyed by the bare tool name coexist for up to 60 days. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs(frontend): state what the tool-call telemetry statuses do not cover `ok` means the tool function resolved, which includes tools that report failure by returning an error string, and a call abandoned mid-execution logs nothing. Also pin that a hallucinated tool name reaches telemetry nowhere. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Guilhem
|
6749015fbf |
fix: audit the icon library against brand guidelines (#10722)
* feat: audit the icon library against brand guidelines
Every icon component checked against its brand's own published guidelines for
correct artwork, current colours, and readability on both app surfaces.
- 127 marks now carry a per-theme pair (text-[#light] dark:text-[#dark]), applied
only where the brand publishes a reversed or dark variant. twMerge where the
component exposes a class prop, so callers can still pass sizing.
- 296 of 304 brand icons record their source in a comment above the <svg>,
including the rule where the brand imposes one (Google forbids recolouring,
Cal.com is deliberately greyscale, Oracle reserves the MySQL dolphin).
- BRAND_COLORS.md is generated from the components, so the table cannot drift
from the code.
- Marks that were unreadable on a surface: 13 -> 1 on dark, 9 -> 4 on light.
The remainder are blocked by trademark terms, not unfixed.
- Wrong artwork replaced where a first-party or CC0 source existed: PayPal is
the real three-colour monogram, Stripe is the bare S rather than an app tile,
gcloud resolves to Google's mark instead of a generic hexagon.
- Concept icons (CACertificate, DbIcon, Webdav, Asset*, Bcrypt) inherit
currentColor instead of hardcoding a colour.
Fixes a cross-component CSS bug: ten icons embedded <style> inside their <svg>.
Svelte only scopes a component's top-level style block, so those were injected as
document-global rules under names like .st0 and .cls-2, which four icons each
defined differently. WindmillIcon renders from the root logged-in layout, putting
.st0 { fill:#ffffff } on every page. Class names are now namespaced per icon.
Adds /kitchen_sink/icons, a gallery rendering every icon on both surfaces at once.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* feat: render brand icons in the surrounding text colour in control strips
A trigger picker mixing lucide glyphs (Webhook, Route, Database) with brand marks
(Kafka, GCP, AWS) read as two sets of controls once the marks became coloured.
Adds an .icon-mono utility that redirects descendant fills to currentColor, applied
by the container rather than passed to the icon. That is what makes it work on every
icon: GoogleCloudIcon has four hardcoded fills, no currentColor and no class prop, so
nothing passed to it could change its colour, and gradient-based marks cannot express
a monochrome variant at all without being redrawn.
- ToggleButton takes a monochromeIcon prop, opt-in since it is used app-wide.
- TriggersBadge, SidebarContent and QuickMenuItem (which backs GlobalSearchModal)
apply it unconditionally: these are uniform lists where one coloured entry among
grey ones reads as an error.
- DropdownV2 gains menuClass, because it portals its menu and a wrapper around the
component cannot style it. CaptureButton passes icon-mono through it.
!important is required because a handful of icons paint through style="fill:…", which
no selector outranks. Stroke is redirected only where one is declared, so shapes
carrying stroke="none" do not sprout outlines. Wrappers use display:contents, so no
layout box is added.
RowIcon is deliberately untouched — table rows keep showing brand colour.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* docs: close out the icon provenance gaps
Sources the 8 icons that had none and settles the 54 records whose author rated
itself below "high" and which no verifier ever reached — the earlier run's
verification stage was killed by a session limit.
46 confirmed as already correct, 11 citations corrected, 3 colours corrected.
Two changes were refuted and reverted by the adversarial pass:
- Mysql: the comment had the colour-to-shape mapping inverted. Rasterising the
first-party asset shows #00758F paints the dolphin and "My" while #F29111 paints
"SQL", not the reverse. The mark renders monochrome here, so nothing on screen
was ever wrong — only the note. Also rescoped the trademark sentence to what the
page literally says.
- AdobeAcrobatSignIcon: a "corrected" citation was rejected on evidence. The agent
claimed the original URL 404s; three fetches returned HTTP 200 with a genuine
Adobe SVG whose stylesheet is .a{fill:#584ccc}. Reverted to the original comment,
which also resolves the one unverified colour change on this branch — #584CCC is
current and first-party confirmed.
AmqpIcon is deliberately left with no brand colour: AMQP is an OASIS protocol, not
a vendor, and amqp.org publishes no palette.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* feat: add icons for 11 resource types that had none
19 hub resource types fell back to a generic Boxes glyph. One agent per brand went
looking for a square vector logomark from a first-party source, with an adversarial
check on everything it produced; 11 landed and 8 correctly came back empty.
Added: beamer, campayn, codat, comapeo_server, klaviyo, matteroom, mollie, motimate,
paychex, terra, vectara. Each records its source, and the components follow the
library's conventions — no <style> block (Svelte does not scope those, which is what
made .st0 leak document-wide), gradient ids prefixed with the component name.
The other 8 keep the fallback, which is the right outcome rather than a failure:
- actimo, adrapid, aero_workflow, matteroom-adjacent niche products publish their mark
only as raster. Upscaled PNGs would look soft beside 300+ vector marks.
- gfw redirects to Global Nature Watch and publishes a wordmark, not a mark.
- leonardoai, localcontexts, weatherapi, webscrapingai serve nothing usable.
No hand-tracing: approximating a mark from a screenshot is invention, not sourcing,
and a wrong logo is worse than the tidy fallback glyph.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* feat: lettermark fallback for reserved marks, and fix the resources table rows
Icons
- Brands that reserve their logo for licensees no longer ship it. BrandLetterIcon draws
the initial in the brand's own colour instead: recognisable, not their mark, and not
invented artwork. Adobe Acrobat Sign and MySQL use it, plus the eight resource types
whose brands publish no vector mark at all.
Adobe: "does not allow the use of its product icons by third parties in their products
or related materials of any kind, except through an Adobe partnership agreement".
On dark the letter inverts to a filled rounded square, because a mid-tone brand colour
chosen to read on white goes dim as a foreground on #2e3441. Where white-on-tile is
also dim, the tile takes a near-black letter instead; light-mode letters are darkened
along their own hue until they clear 3:1. Every pair was measured, not eyeballed.
- Google Docs was drawing a generic monochrome document glyph while carrying a comment
claiming Google's colours. Replaced with Google's own 192px product icon.
- Azure was drawn monochrome, justified by a comment citing Microsoft's rule against
distorting the mark — which drawing it monochrome is. Replaced with Microsoft's own
logo_azure.svg. Their terms say to use the icons "as they would appear within Azure";
permitted use is diagrams, training and documentation, which is recorded in the file.
- Adobe Acrobat Sign's artwork was a geometric "A" plus a squiggle, not Adobe's ribbon
swirl. Moot now that it is a lettermark, but the mark was wrong.
- Gradient, mask and clip ids in the new artwork are namespaced per icon; ids are
document-global and collide the same way the .st0 class names did.
Resources tables
- Description cells are a fixed two lines: min-h floors short ones, line-clamp ceilings
long ones, so every row is the same height. Full text on hover via title.
- Widened to 30rem (84 chars/line) and vertically centred. The clamp needs
display:-webkit-box, which stacks lines from the top, so the span sits in a
flex items-center wrapper rather than carrying the height itself.
- w-full min-w-0 max-w-[30rem] instead of a fixed w-96, so a narrow viewport shrinks the
column and truncates rather than forcing the page to scroll sideways.
- The actions column loses its border-l separator and right-aligns the "Shared globally"
badge, matching the rows that show buttons.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: icon-mono filled lucide outlines and missed currentColor brand marks
Two bugs in the monochrome utility, both from the fill rule being too blunt.
- Lucide icons are outlines: fill="none" with stroke="currentColor" and no fills on
their children. Forcing fill on every descendant overrode that none and turned each
glyph into a solid blob. The filled case is now scoped to svgs that do not declare
fill="none", and svgs that do only get children redirected if they declare a real
fill of their own — so a brand mark drawn as an outline still works.
- Brand marks that paint with currentColor carry their own text-[#hex] class, so
redirecting fills left them branded: MQTT stayed #660066, NATS #375C93. The svg now
inherits the container's colour, which is what actually makes them monochrome.
Also wires the sidebar's trigger section, which was never covered: those links render
through MenuLink, not the sub-item block that had the class.
Verified in the browser across all five shapes an icon can take — lucide outline,
hardcoded fill, currentColor plus brand class, outline root with filled children, and
inline style="fill:#..". Lucide keeps fill:none and a grey stroke; the rest follow the
container.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* refactor: dedicated monochrome trigger icons instead of a CSS override
Reverts the trigger surfaces to the icons that were there before the brand-colour
audit, as ./icons/triggers/ variants. A trigger picker lists brand marks beside lucide
glyphs (Webhook, Route, Database), so a coloured mark reads as a different kind of
thing rather than a peer.
Ten variants, restored from main where they were already monochrome: Kafka, NATS, MQTT,
AMQP, AWS, Azure, Nextcloud, Google, GitHub. Google Cloud is the exception — main's copy
is a greyscale rendition rather than currentColor, so it is rebuilt from the current
four-colour artwork with the fills dropped.
Separate files rather than the CSS override that was there, because coercion cannot work
in general: forcing fills to currentColor breaks lucide's outline icons, which are
fill="none" with a stroke, and marks that set their own text-[#hex] class ignore a fill
rule entirely. Both bugs were live. The .icon-mono utility, ToggleButton's monochromeIcon
prop and the DropdownV2 menuClass pass-through are gone with it.
index.ts documents which folder to use where: ./triggers/ for trigger surfaces, the
full-colour mark for the resource picker, AppConnect and docs, and keep the two in sync.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: trigger pages and global search still used the colour brand marks
The ToggleButtonGroup on each trigger page pairs a brand icon with a lucide Code
glyph, so GCP Pub/Sub rendered Google's four-colour mark next to a monochrome one.
Kafka, NATS, MQTT and the rest had the same wiring; they were just less obvious
because their marks are near-monochrome already.
Repoints all seven trigger pages and the global search nav entries at the
./icons/triggers/ variants. RowIcon is left on the full-colour marks: table rows
show brand colour by design.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: restore the greyscale GCP trigger icon, and show variants in the gallery
The trigger variant had been flattened to currentColor, which collapses Google's cloud
into one flat silhouette and loses the tonal steps that give it shape. The pre-audit
icon was greyscale, not monochrome — #B0B0B0 / #D0D0D0 / #E0E0E0 / #FFFFFF — so it is
restored verbatim from main.
Also globs icons/**/*.svelte in /kitchen_sink/icons so trigger variants render next to
the full-colour marks they shadow, labelled by folder. Comparing the two is the thing
this page was missing.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: flow trigger dropdown rows use the desaturated marks too
The flow-graph badge menu still rendered the full-colour brand marks next to
lucide glyphs. Route both dropdowns through triggerIconMapMono: the badge
itself keeps the colour mark, only the rows it opens change.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* feat: rank resource-type search results by best match
Searching the description is what makes `gdrive` findable as "google", but it
also means "google" matches a dozen types that only mention the product in
passing. Rank a match on the type's own name above any description match, and
break ties on where the match starts, so `googleai` leads and a description
opening with "Google OAuth token..." beats one mentioning Google halfway
through.
Applied to all three resource-type searches: the Resource Types tab (whose bare
term also only searched the name until now), the add-resource drawer, and the
schema-narrowing picker.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: trigger pages and global search show the full-colour marks
The desaturated variants belong to the two dense lists that sit beside lucide
glyphs -- the sidebar trigger list and the capture dropdown. Everywhere else a
brand mark stands on its own and should be the real one: the per-kind trigger
pages, the command palette, the capture table and the chat tool cards. Records
the rule in icons/index.ts so the next caller picks the right folder.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: address review round on the icon and resource-type work
- AppConnectInner went back to listResourceTypeNames for the list: /resources/type/list
is not on the public app domain's route allow-list, so a published app's resource
picker 403'd and, because the throw left connectsManual unset, stayed empty on every
retry. Descriptions now load best-effort behind it.
- Dropped DropdownV2's menuClass: nothing passes it; the flow-graph badge menu styles
melt's Menu, which has its own.
- icons/index.ts named two surfaces for the desaturated variants; there are four, and
the flow-graph badge and the menu it opens differ. Dropped the stale GCloudIcon note.
- GoogleCloudIcon takes width/height again: generic call sites resolve it through
APP_TO_ICON_COMPONENT and pass no size, so gcloud rendered at 16px after the remap.
- The path explainer is one ResourcePathHint component instead of the same copy twice.
- BRAND_COLORS.md recorded Ansible, Datadog, Deno, DeepL and Toggl as fixed; each
publishes a second artwork swapped in by class, so their dark hex and ratio were
wrong. Header no longer claims a generator that isn't in the repo.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: drop the duplicate gcloud icon and unblock the connect list
GCloudIcon.svelte was rewritten into the same four-colour mark as
GoogleCloudIcon.svelte and nothing pointed at it any more, so it was two files
drawing one logo waiting to drift apart.
The description fetch also sat on the critical path: the "Others" list showed
skeletons until a request for every type's full schema returned -- one that a
published app is guaranteed to get a 403 on. It now runs unawaited, and search
re-ranks when the descriptions land.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: AwsIcon and GoogleIcon take size again
The audit narrowed both to width/height with a 24px default, but every dynamic
call site passes size — RowIcon, the flow trigger badges, ToggleButton, global
search, the chat tool cards, the native-trigger page — so the SQS and Google
marks rendered at 24px wherever a smaller size was asked for. Both take size
again, keep width/height for the call sites that use those, and accept a class
so RowIcon's grey still applies.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: row-strip brand marks keep their colour
RowIcon greyed five of its seven brand marks with text-gray-400 while gcp and
azure rendered in colour. Now that AwsIcon accepts a class, the grey took its
wordmark but not its hardcoded #FF9900 smile, so the SQS row came out half
grey, half orange.
The rule this branch settled on is that only the four trigger menus desaturate;
a table is not one of them. Dropping the class from all five makes the strip
agree with the gcp and azure rows beside them, and with the lucide glyphs
staying grey since they carry no brand colour to keep.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
hugocasa
|
8492b4b4ba |
chore: prove scratch file ops per command segment (#10744)
* fix(agents): prove scratch file ops per command segment Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs: describe the checkout root in the scratch guidance Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(agents): close two auto-allow holes in the scratch guards Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(agents): keep redirects and chained writes off the allow path Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(agents): never prove a command carrying a substitution or relative cd Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(agents): treat sibling checkouts as separate roots Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(agents): prove where a directory-form cp or mv actually lands Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(agents): leave directory-form cp and mv unproved Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs: state the one-write-per-line rule in the scratch guidance Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs: prefer Edit/Write over shell edits in agent guidance Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(agents): stop a failed cd from hiding the directory form Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor(agents): state the glob and cd rationale once Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Ruben Fiszel
|
6783a396b1 |
fix(api): document cache_ignore_s3_path on the Script read schema (#10742)
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Ruben Fiszel
|
1fa3bf3b29 |
fix: show runtime-detected assets in a run's Assets tab (#10738)
* fix: show runtime-detected assets in a run's Assets tab * fix: address review nits on run assets tab * fix: cap the run assets list and report when it is cut * fix: cap run assets by asset, not by row |
||
|
Ruben Fiszel
|
5d7881beb8 |
test: settle the backlog before the capped audit-export drain (#10737)
* test: settle the backlog before the capped audit-export drain * chore: update ee-repo-ref to bd4de74eb37b32a2b6c7c69f6dedac031ef8436b This commit updates the EE repository reference after PR #730 was merged in windmill-ee-private. Previous ee-repo-ref: b5a5f9114df26088cfe976d91f10e55ba8bfcaa6 New ee-repo-ref: bd4de74eb37b32a2b6c7c69f6dedac031ef8436b Automated by sync-ee-ref workflow. --------- Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com> |
||
|
Ruben Fiszel
|
ce71756c89 |
chore(main): release 1.791.0 (#10718)
* chore(main): release 1.791.0 * Apply automatic changes --------- Co-authored-by: rubenfiszel <275584+rubenfiszel@users.noreply.github.com>v1.791.0 |
||
|
Ruben Fiszel
|
343ce6e143 |
fix: derive a raw app's policy on deploy, and default an omitted execution_mode (#10733)
* fix: default an omitted app policy execution_mode to publisher Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs: drop stale comments claiming execution_mode is required Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: derive a raw app's policy on deploy instead of trusting the caller's Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * chore: pin the ee ref to the companion branch Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: vendor the raw-app policy derivation into the bundle job Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs: note the vendored raw-app policy bundle Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: derive the policy on a value-only raw-source update too Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: reject raw-app runnables whose shape yields an unusable grant Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: cache the new policy query and tighten raw-app runnable validation Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: let the policy bundle drift guard survive a CRLF checkout Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * chore: update ee-repo-ref to 23431f5cf1d627051ded89111bbf2e301e9db456 This commit updates the EE repository reference after PR #729 was merged in windmill-ee-private. Previous ee-repo-ref: 0bdf8818fa115ad6b0d14f3117a18e8a580cce4d New ee-repo-ref: 23431f5cf1d627051ded89111bbf2e301e9db456 Automated by sync-ee-ref workflow. --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com> |
||
|
Ruben Fiszel
|
5972a1ca06 |
chore(cli): pick up shared-utils 1.0.13 (#10736)
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Ruben Fiszel
|
39f0542b2f |
refactor(frontend): keep the shared-utils bundle free of UI code (#10735)
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Ruben Fiszel
|
f4f2dd5ece |
chore(frontend): unbreak the shared-utils library build (#10734)
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Ruben Fiszel
|
b17fdab8ff |
fix: compile resource types with no properties instead of throwing (#10730)
* fix: compile resource types with no properties instead of throwing * fix: keep property-less resource types in the editor RT namespace |
||
|
Ruben Fiszel
|
6b5b9f72d4 |
fix: type s3-streamed columns that are all-null in the inference sample (#10728)
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
 AlexRV12
|
fd9295a58e |
feat(copilot): let plan mode draw, but never write the plan (#10725)
* fix(copilot): validate the version an approval stamps * feat(copilot): let plan mode write artifacts, but never the plan * feat(copilot): tell plan mode it may keep notes, not rewrite the plan |
||
|
Guilhem
|
66bffaa60d |
feat: add empty state cards to list pages (#10726)
* feat: add empty state cards to list pages
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: animate trigger drawers on first open
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: distinguish filtered-empty schedules, reuse the rAF helper
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: hide the header create button while the empty state offers it
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* Revert "fix: hide the header create button while the empty state offers it"
This reverts commit
|
||
|
Ruben Fiszel
|
05eba6c9ab | fix: include delete_after_secs in script deploy payload (#10731) | ||
|
Ruben Fiszel
|
66e3790da4 |
docs: announce we are not seeking outside contribution (#10724)
* docs: announce we are not seeking outside contribution Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs: point big ideas at the feature request template Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
 Pyra
|
ab3c0206d7 |
fix: support @typechecked decorator in Python relative imports (#8495)
WindmillFinder's ModuleSpec lacked origin, so __file__ was never set on loaded modules. inspect.getfile() then raised "is a built-in module", breaking typeguard's @typechecked and anything else that introspects module source. Use spec_from_file_location() which sets origin correctly. Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com> Co-authored-by: hugocasa <hugo@casademont.ch> |
||
|
Ruben Fiszel
|
010d67e07f |
chore(main): release 1.790.1 (#10712)
* chore(main): release 1.790.1 * Apply automatic changes --------- Co-authored-by: rubenfiszel <275584+rubenfiszel@users.noreply.github.com>v1.790.1 |
||
|
Ruben Fiszel
|
64d78b4db1 |
fix: fall back to polling when a proxy mutes the job SSE stream (#10716)
* fix: fall back to polling when a proxy mutes the job SSE stream * fix: do not charge deliberate no-logs sse restarts to the retry budget |
||
|
Ruben Fiszel
|
529e960629 |
perf: cap resource content sent to the search modal (#10714)
* perf: cap resource content sent to the search modal Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: address review — fence the LATERAL, flag partial search, add cap test Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: pluralize the truncation notice and link the cap to its openapi doc Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Ruben Fiszel
|
0258f3f81b |
perf: unblock workers before the API router is built (#10711)
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Ruben Fiszel
|
944ad1083a |
chore(main): release 1.790.0 (#10699)
* chore(main): release 1.790.0 * Apply automatic changes --------- Co-authored-by: rubenfiszel <275584+rubenfiszel@users.noreply.github.com>v1.790.0 |
||
|
hugocasa
|
effdcd9915 |
fix: recover from a refused mcp read assertion, drop stale discovery (#10710)
* fix: recover from a refused mcp read assertion, drop stale discovery Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: drop the stale listing from the raw error, not the bounded payload Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
hugocasa
|
3468cb68b1 |
fix: drop sampling params on Claude models that reject them (#10708)
* fix: drop sampling params on Claude models that reject them Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs: scope the sampling-param claim to what was probed and split the bedrock test Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * test: build the disable body through the resolver instead of asserting a rejected shape Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: use the Gemini 3.1 Pro id that actually resolves Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: Bedrock Sonnet 5 cannot disable thinking, unlike the native API Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Ruben Fiszel
|
c9ddddda1b |
log the settings a failed read left unapplied (#10709)
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Ruben Fiszel
|
d97f380c87 | test: stop stranding sqlx pool permits in run_in_isolated_thread (#10707) | ||
|
hugocasa
|
ee533273dd |
fix: confine jobs:run tokens to the jobs of the runnables they may start (#10635)
* fix: confine path-scoped jobs:run tokens to their runnable's jobs Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: project singlestepflow onto its runnable and confine kind-only run scopes Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: keep every by-id job read reachable by a jobs:run token Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: whitelist the dbt and wac-approval by-id job reads for run tokens Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: let an apps:run scope satisfy job-read confinement for that app's runs Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: apply run-scope confinement on top of the approval-token read bypass Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: confine the resume-secret job reads to the run scope as well Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
hugocasa
|
3f07a1a803 |
feat: let the global AI chat call connected MCP servers as the user (#10656)
* feat: let the global AI chat call connected MCP servers as the user Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: address review findings on the chat MCP tools Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: connect MCP servers from a predefined list in chat and agent steps Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: show the OAuth redirect URL in the instance connect settings Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: clarify the OAuth redirect URL copy in instance settings Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: match the instance settings warning style and drop the redirect tooltip Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: use the standard warning alert for the redirect url mismatch Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: correct the GitHub token guidance in the MCP registry Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: warn when an OAuth connect lacks the scopes an MCP server needs Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: request the connect's scopes when the oauth popup is opened directly Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: connect an oauth-app MCP server without leaving the panel Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: seed connect scopes from the instance config only Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: make the chat use only the MCP servers you turn on Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: align the MCP connect UI with the design system * feat: make a pasted url the default way to connect an mcp server * feat: show provider icons on the suggested mcp servers * fix: make both mcp sign-in paths behave the same and stop reloading on toggle * fix: clarify the mcp tool step's server field and drop its info alert * fix: name the mcp resource in the tool step and move the transport note into the connect box * fix: drop the redundant description on the mcp resource field * fix: make the mcp connections trigger icon-only * fix: scope enabled mcp servers to the account and address review nits * fix: wait for connect scopes and create session connections in the operating workspace * feat: move mcp connections into the chat's plus menu and fix review findings * fix: show mcp servers as checkboxes so off reads as a state * feat: give menu rows an on/off switch and use it for mcp servers * fix: lead the mcp menu rows with the switch * feat: keep the menu open while toggling and simplify the connect card * fix: ask for the server before the credential in the connect card * fix: show one credential path at a time in the connect card * fix: label the path field and move token guidance into its tooltip * fix: open straight into connect and keep the server menu scannable * feat: warn when an mcp connection lands outside your own space * refactor: require the workspace on the mcp connect components and rename the oauth child * fix: replace the oauth variable on reconnect and bound every mcp result * feat: show a connected server's provider icon in the connections list * feat: resolve mcp provider icons from the url and clarify the path field * style: align the mcp connect card with the design system surfaces * style: drop the redundant oauth support line and name the scopes oauth scopes * feat: keep the mcp connect card open in the connections drawer * feat: preopen the mcp connect card under the agent step resource picker * feat: resolve a typed mcp url to its registry entry and describe the token field * style: name both mcp connect actions connect * style: name the mcp oauth actions connect with the provider * style: say in the path description what the connect action will save * style: name the resource type in the mcp connect path description * feat: cache mcp provider icons and confirm disconnect in a modal * fix: keep the mcp menu switches live and the disconnect modal above the drawer * style: fall back to the plug icon in the mcp menu rows * fix: never destroy a foreign variable or resource when connecting an mcp server * fix: prove a token variable is ours before writing it and bound mcp search failures * fix: pin an mcp oauth popup to the target it was opened for * fix: bind an mcp credential to the server and popup it was requested for * fix: bound mcp tool calls with a deadline and drop stale server listings * fix: keep the disconnect confirmation handler returning void * fix: tie the mcp tool cache to the resource revision and the grant to its scopes * fix: verify mcp read-only server-side, keep oauth connector mounted Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
hugocasa
|
53eb94659b |
feat(telemetry): extend feature-usage tracking beyond AI features (#10681)
* feat(telemetry): extend feature-usage tracking to long-tail features Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs: describe telemetry as product feature usage rather than AI usage Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor(telemetry): trim disclosure copy and drop unused pick origin Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(telemetry): count trigger fires per run and key hub picks from hub data Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(telemetry): slugify hub keys and order both writers' upserts Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(telemetry): key native trigger adoption by service so it matches fires Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * chore: update ee-repo-ref for native trigger adoption fix Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor(telemetry): move feature-usage collection into the ee crate * docs: point feature-telemetry at the moved registry and rust writer * docs: correct the trigger-fire gate comment to match measured step counts * docs: put the private-build caveat on the verification step * chore: update ee-repo-ref to f079db9e7962a413b349c4ff8036080894f30771 This commit updates the EE repository reference after PR #725 was merged in windmill-ee-private. Previous ee-repo-ref: 055adb80416f9339c9a28ae7fbaeadad30d74959 New ee-repo-ref: f079db9e7962a413b349c4ff8036080894f30771 Automated by sync-ee-ref workflow. --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-authored-by: Ruben Fiszel <ruben@windmill.dev> Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com> |
||
|
hugocasa
|
98bacab907 |
refactor: combine the per-minute counters onto one shared helper (#10687)
* refactor: combine the per-minute counters onto one shared helper Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: keep dashmap in windmill-store for the azure devops token cache Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: name the sweep counter for what it counts Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
hugocasa
|
bd5b3ea779 |
fix: send sage_intacct oauth client credentials in the request body (#10685)
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Guilhem
|
1a606b1856 |
fix flaky sessionState IndexedDB hydration race (#10692)
* test: fix flaky sessionState IndexedDB hydration race * test: fold logout into the login barrier helper |
||
 Alexander Petric
|
b5510333ea |
fix(groups): replace instance-group delta-patching with a state-based reconciler (#10686)
* fix(groups): replace instance-group delta-patching with a state-based reconciler Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TYi5MGYwUL6gjYMgQVY2Yx * fix(groups): follow instance-group renames through workspace references Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TYi5MGYwUL6gjYMgQVY2Yx * fix(groups): preserve historically-orphaned instance-group members on upgrade Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TYi5MGYwUL6gjYMgQVY2Yx * fix(groups): preserve retained-group orphans too in the upgrade migration Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TYi5MGYwUL6gjYMgQVY2Yx * test(groups): exercise the orphan-preservation migration; strip refs before converting Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TYi5MGYwUL6gjYMgQVY2Yx * test(groups): pin the migration's strip-before-convert order Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TYi5MGYwUL6gjYMgQVY2Yx * fix(groups): make reconciliation the last locking step in every mutation path Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TYi5MGYwUL6gjYMgQVY2Yx * fix(groups): make the workspace advisory lock first in the lock hierarchy Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TYi5MGYwUL6gjYMgQVY2Yx * fix(groups): lock workspaces before membership writes in single-user paths Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TYi5MGYwUL6gjYMgQVY2Yx * fix(groups): use the instance_group row as the group-level mutex Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TYi5MGYwUL6gjYMgQVY2Yx * fix(groups): take an exclusive instance_group table lock in overwrite_igroups Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TYi5MGYwUL6gjYMgQVY2Yx * chore: update ee-repo-ref to af02d6bce55512b65c56adcbf69a8e15cd124d23 This commit updates the EE repository reference after PR #726 was merged in windmill-ee-private. Previous ee-repo-ref: ec2feac82636869731666e5c6578b6c078e9aeb2 New ee-repo-ref: af02d6bce55512b65c56adcbf69a8e15cd124d23 Automated by sync-ee-ref workflow. --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com> |
||
|
hugocasa
|
68fc7825bb |
fix: refresh AI provider model defaults and capability metadata (#10690)
* fix: refresh AI provider model defaults and capability metadata Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: send explicit thinking disable for Claude and cap Opus 4.1 output Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: resolve mistral-medium-latest window and OpenRouter Claude 5 off Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: cover au. bedrock geo and Fable 5 caching, revert unverified mistral ladder Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: scope the Anthropic explicit disable to models that think by default Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: translate the reasoning off sentinel on the backend Anthropic path Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: translate the reasoning off sentinel on the Bedrock Converse path Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: share the reasoning off sentinel and make its translation testable Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
AlexRV12
|
850b028778 |
feat: advertise the pinned artifact version in get_preview_status (#10691)
* test: let global evals seed the session's preview tabs Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: advertise the pinned artifact version in get_preview_status Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * test: reject ambiguous preview-tab and artifact eval fixtures Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Guilhem
|
60c5ad252a |
fix: keep a resource's linked secret reference in sync while renaming (#10693)
* fix: keep a resource's linked secret reference in sync while renaming Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: guard null resource args when renaming Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Ruben Fiszel
|
e6e2e53e97 |
fix(agents): stop the scratch-dir guards prompting on quoted text (#10703)
* fix(agents): stop the scratch-dir guards prompting on quoted text * fix(agents): keep prompting past wrapper flags and quoted heredoc markers * fix(agents): only treat a line-ending delimiter as a heredoc opener * fix(agents): refuse a heredoc opener whose redirect carries a quote * fix(agents): stop the wrapper scan at a quoted word instead of a word count * fix(agents): scan a wrapper's operands to the end of the segment * fix(agents): scan a heredoc body that is piped into a shell * fix(agents): only treat a quoted, unexecuted heredoc body as data * fix(agents): split separators before looking for the shell running a heredoc * fix(agents): require a reading consumer before treating a body as data |
||
|
Ruben Fiszel
|
0fc74dec5f |
fix(ci): use random delimiters for untrusted multiline workflow values (#10706)
* fix(ci): use a random delimiter for the review prompt env var * fix(ci): use a random delimiter for the review command extra_prompt output |
||
|
Ruben Fiszel
|
30f5d2e766 |
perf: declare a settings pass instead of reading one setting at a time (#10698)
* perf: read global_settings once per settings-load pass `initial_load` reads several dozen settings back to back, one `SELECT value FROM global_settings WHERE name = $1` each: 50 serialized round trips before a worker is ready, 32 before a server is. On localhost that is ~20ms and invisible; against a real database it is 50x the RTT per process start, which `EXIT_AFTER_N_JOBS` turns into a per-job cost. `with_global_settings_snapshot` reads the whole table (12 rows on a typical instance) into a tokio task-local, and `load_value_from_global_settings` serves from it. Scoping it to the task is what keeps the single-setting reload paths correct: a `notify_global_setting_change` event for one key runs outside any scope and still reads the database, so a live settings change reaches a running worker as before. Agent workers hold an HTTP connection with no snapshot to take and are unchanged. `load_smtp_config` and `reload_custom_tags_setting` had their own inline copies of the same query; they go through the shared loader so they land in the snapshot too. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs: state the snapshot contract on the reader and the query `load_value_from_global_settings` is called from ~10 crates and one of them writes a setting then immediately re-reads it through `reload_custom_tags_setting`; say on the function itself that a scope, when one is installed, serves the read and leaves `db` unused. The query comment claimed the table is a handful of rows. It is not bounded that way: `workspace_dependencies_map_rebuilt:<workspace_id>` adds a row per workspace and never removes it. Those dynamically named rows are also why the snapshot fetches the whole table instead of the wanted names, so state that as the reason. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: bound the settings snapshot and keep it out of two reads Three review findings, all real: The snapshot fetched the whole table, which is not bounded by the settings that exist: `workspace_dependencies_map_rebuilt:<workspace_id>` adds a row per workspace with no cleanup path, and no settings pass reads one. It now fetches only statically named rows, and reads of a `<prefix>:<id>` name skip the snapshot and go to the database. Correctness does not rest on that naming convention — a colon-free dynamic name would simply be in the snapshot and still answered correctly — only the bound does. A snapshot query that failed inside an enclosing snapshot awaited the body bare, so its reads were served by the outer snapshot rather than falling through as documented. The task-local carries an explicit bypass state and the failure path scopes it. `reload_jwt_secret_setting` decided whether to generate-and-upsert the JWT secret from a snapshot-served read, so a replica booting alongside another could overwrite the secret it had just generated and invalidate its tokens. That read goes through the new `load_value_from_global_settings_fresh`. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: keep the snapshot query on the primary-key index `name NOT LIKE '%:%'` bounded the rows returned but not the work: a leading wildcard cannot use the index, so Postgres read every row anyway. Against 50k dynamically named rows it plans as a seq scan of 516 buffers whether or not seqscans are enabled — and worker connections disable them, so the plan was one the query shape forbade rather than one the planner chose. `name = ANY($1)` over an explicit list plans as a bitmap index scan, 7 buffers, bounded by the listed names rather than by table size. That list is also exactly the set the snapshot may answer from, so a name outside it falls through to the database instead of reading as unset: listing a setting is a performance choice, never a correctness one, which is what keeps the list safe to maintain by hand. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: declare a settings pass instead of reading one setting at a time Replaces the prefetch-list snapshot with a pass the call sites build themselves. `SettingsPass` collects the reads `initial_load` will make as `(name, applier)` pairs, fetches them together, then replays the appliers in declaration order. Declaring is what makes the batch exact. The same `if server_mode` / `if *CLOUD_HOSTED` / `cfg` branches that used to guard a read now guard a declaration, so the fetch asks for what this process needs and nothing else, and there is no list of setting names to keep in sync with anything. Ordering is preserved end to end: appliers run in the order they were declared, and non-setting work in the middle of the sequence keeps its place as a step, so nothing moves and nothing runs twice. Steps that need several settings at once take them together. The batch distinguishes three states where a per-setting read only ever produced two at a given call site: - a value, - genuinely unset, which several settings must see in order to restore a default when the setting is cleared, - could not be read, which must leave the in-memory value alone. Collapsing this into "unset" would let one failed query reset workspace fairness and the queue caps across a cluster. Over HTTP the reads go out together rather than sequentially, so an agent worker's settings load costs one round instead of ~36, with no new endpoint. A setting an agent may not request still resolves to unset, as the per-setting call returned for it. `reload_*` keeps working per setting for the notify path, sharing its apply half with the pass. The wrappers no caller was left using are dropped. worker startup: 50 queries -> 2 (the batch, and jwt_secret which stays its own read so the pass cannot sit between reading it absent and upserting a replacement over another replica's). Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: run the pass's non-setting steps in declaration order too Review round found the settings pass had a gap: the reads were declared but the work interleaved between them still awaited inline, so it all ran before `pass.run` applied anything. `manage_audit_partitions` therefore saw `AUDIT_LOG_RETENTION_DAYS` at its compile-time default rather than the configured value, and dropped every partition past that default. An instance keeping 30 days on CE lost the 14-to-30-day band on startup and on every full-reload tick. The `STORE_AUDIT_LOGS_S3` export anchor had the same cause: the gate read `false` before the setting applied, so an env-var-enabled export never anchored and its first tick skipped the rows committed before it. `action` exists so a step keeps its place in the sequence; every remaining inline await is now one, which fixes both and leaves no phase where a read can observe a value the pass has not applied yet. Two more from the same round: A batch that fails as a whole now falls back to per-setting reads. Skipping every applier preserves known-good state on a reload tick, but a starting process has none, and would have run on compile-time defaults until the next full reload twelve hours later. `FORCE_RUBY_REPOS` is honored again: the batched url-list path parsed without the `FORCE_` check its per-setting counterpart applied, so the override was silently dropped. `load_setting_value` never had one, so the third helper was never affected. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: declare the object-store and worker-config steps in the pass too Two awaits were left running ahead of `pass.run`, so the settings they read were still at their compile-time defaults. The object-store reload is the one that matters: an AWS OIDC store mints its first token against an issuer built from `BASE_URL` (`oidc_ee.rs`), and with `OTEL_ENVIRONMENT` set nothing loads that before this pass does, so the store signed with the unset default, left `OBJECT_STORE_SETTINGS` empty and fell back to the ten-second retry while startup carried on. `reload_worker_config` calls `store_pull_query`, which reads the workspace fairness knobs. It happened to converge because the enabled flag re-stores the query when it changes, but it was reading defaults on the way there. Both are steps now, which is also what the earlier fix should have covered: the only await left outside a step is `pass.run` itself. Also from the same round: `fetch_settings_batch`'s doc comment had been stranded on the helper inserted above it, and the batch-failure fallback re-ran the same reads on an agent worker, where the batch already is the per-setting read. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs: point the setting-loader docs at functions that still exist `reload_setting` went with the other wrappers no caller was left using, but two doc links still referenced it. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: decide the jwt secret in sql so the read can be batched `reload_jwt_secret_setting` generated a secret whenever its read came back absent or unparseable, and upserted it unconditionally. Two replicas booting against an empty row therefore each installed their own and rejected each other's tokens, and the same happened on a running cluster whenever the row was deleted or set to a non-string. Keeping the read next to the write kept the window narrow but never closed it, and it was the reason this one setting could not go through the settings pass. `get_or_create_jwt_secret` puts the decision in the statement instead: INSERT ... ON CONFLICT (name) DO UPDATE SET value = EXCLUDED.value WHERE jsonb_typeof(global_settings.value) <> 'string' RETURNING value First writer wins, a usable secret is never overwritten, and an empty RETURNING is how a caller learns another process's secret stands. The `WHERE` also keeps a normal startup from writing at all, which matters because `notify_global_setting_change` fires on every write to this table and an unconditional upsert would have made each start trigger a cluster-wide reload. Because the statement decides rather than the caller's read, a stale value is harmless and `jwt_secret` is now an ordinary declaration. Worker startup is a single batch round. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: keep a failed read from dropping a FORCE_ override or clearing a setting Two ways a read that did not succeed was being treated as an answer. A `FORCE_` override used to be checked before the read, so a failed read could not affect it. Moving that check into the parser put it behind a value arriving, and a failed read skips its applier, so a forced private registry fell back to the public index and a forced `settings.xml` was deleted from disk by the Maven step that follows it. Forced settings are declared as steps with no read now: the override outranks the database, so there is nothing to fetch and nothing to lose when a fetch fails. The setting loaders were passing `v.ok().flatten()` to their appliers, which turns a database error into "unset". Most appliers ignore `None`, but `apply_tag_per_workspace_workspaces` clears the workspace whitelist with it, making every workspace eligible for per-workspace tags, and `apply_fork_workspace_tag_append_fork_suffix` stores `false`. Both are also reached from the notify handlers, so a blip during a reload changed routing for the cluster. They take `?` now, as the code they replaced did by leaving the error arm empty, and the other five are converted with them so an applier that later grows a `None` branch cannot inherit the problem. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: route hub_api_secret through the FORCE-aware declaration `HUB_API_SECRET` lives in an `ArcSwap` rather than an `Arc<RwLock<_>>`, so it could not use `option_setting` and was declared by hand with a bare `setting` plus `parse_option_setting_value` — which is exactly the path that skips the `FORCE_` handling, so a failed read still dropped `FORCE_HUB_API_SECRET`. The rule now lives in `option_setting_with`, which takes the store closure and leaves `option_setting` a wrapper over it, so a setting held in something other than an `RwLock` reaches it too rather than having to reimplement it. The three remaining hand-written parses are `parse_setting_value`, which has no `FORCE_` handling to miss: `load_setting_value` never had the check either. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Ruben Fiszel
|
633d7bcb2e |
feat: add trigger_history table with source tracking (#10696)
* feat: add trigger_history table with source tracking Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: gate trigger history reads on scopes and harden its writers Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: filter trigger history scopes in SQL and match the cleared-handler diff Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: record a trigger restore from the trashbin in its history Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: record bulk http trigger creates and document the recording boundary Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: lock the trigger row when capturing its history preimage Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: only record an auto-disable that actually flipped the schedule Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * chore: state the auto-disable invariant once instead of at four call sites Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: render trigger history changes as a structured field diff Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: make a server-initiated disable atomic with its history row Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs: note that the auto-disable savepoint takes no pool connection Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs: note the flow fallback is the last chance to disable Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: never leave a trigger enabled because its history row failed Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: retry the disable history row instead of dropping it on first failure Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: use the design-system Button for the change-value expander Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: hold the trigger row lock across its disable history row Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: keep the history-loss alert out of the listener cancellation race Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: read the history workspace through the trigger-workspace seam Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Ruben Fiszel
|
6d03784d4b |
fix: keep non traffic-serving processes out of coordinated restarts (#10694)
* fix: key server_heartbeat row on hostname so restarts reuse one row Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * chore: trim announce_server_started doc to the durable constraints Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: only traffic-serving processes take part in coordinated restarts Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * chore: name every non traffic-serving mode in the restart-gate comments Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * chore: narrow the restart-gate comments to claims that hold Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Ruben Fiszel
|
878b8ef4c4 |
perf: cache resolved python interpreter path across worker restarts (#10701)
* perf: cache resolved python interpreter path across worker restarts Every worker process start spawned two `uv python find` subprocesses to re-discover an interpreter path that had not changed, and every python job spawned one more. The resolved paths are now memoized in a small JSON file next to PY_INSTALL_DIR, which outlives the process, so a restarted worker (notably under EXIT_AFTER_N_JOBS) reuses what the previous one resolved. An entry is only served when the uv binary is the same one that produced it and the interpreter is still on disk; otherwise it falls through to a real `uv python find`. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: address review findings on the python path cache - resolve uv through PATH on windows, where `metadata("uv")` looked in the worker's current directory and silently disabled the cache - stat uv with tokio::fs instead of blocking the runtime, and compute the identity once per resolution instead of once per read and twice per write - store one file per version instead of a shared map, so workers resolving different versions concurrently cannot drop each other's entry Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: keep the windows uv PATH probe off the async runtime The lazy static resolving uv through PATH stats candidate entries synchronously, so its first use is moved onto a blocking thread. Also records why an entry keyed on a minor-only version does not pin a patch: uv answers such a request with its minor-version link and re-points it on a patch install, so the memoized path follows the upgrade. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Ruben Fiszel
|
0a40b3806f | fix(agents): let the scratch-dir hooks own their permission prompt (#10702) | ||
|
Ruben Fiszel
|
578d5e9a7d |
perf: back off the interactive worker shell under EXIT_AFTER_N_JOBS (#10700)
* perf: back off the interactive worker shell under EXIT_AFTER_N_JOBS Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * chore: address review nits on the shell backoff docs and periodic warning Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * perf: only give the worker shell its sub-second cadence during a live session Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
AlexRV12
|
caa189868c |
feat(ai-sessions): add plan mode (#10057)
* feat(sessions): let an opener name the artifact version to show A tab already remembers the version a reader pinned, and re-pointing it keeps that pin. Plan mode needs the two intents that leaves out: a plan card scrolled up the transcript wants the version it proposed, and a plan going up for approval wants the current text with no pin at all. `ArtifactVersionTarget` is those two alongside the existing one: a number, `'latest'`, or omitted. Omitted still cannot double as `'latest'` — every artifact tool re-opens the document it just wrote, so taking that as a request to move would yank a reader out of the version they chose on every edit the agent makes. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat(copilot): add the plan-mode gate and tag plan-mode-safe tools Plan mode is a read-only posture, so something has to decide which tools it may still run. `Tool.planModeSafe` is that tag, and processToolCall fails closed on it: untagged means mutating means blocked. Deriving it from `requiresConfirmation` was not an option — unconfirmed mutating tools exist, and a posture that leaks one is not a posture. The gate runs twice per call. Before `validateBeforeConfirmation`, so a validator cannot reach out while planning; and again after the confirmation wait, because plan mode can be entered while a mutating tool's card is already pending, and that approval must not carry it through. Arguments are read one field at a time rather than through a parse of the whole call. `change_note` is optional and cosmetic, and a model that sends it as `null` would otherwise fail the object parse and take the plan down with it — the user being told there was no plan to approve, which is false. Also here, because refusing a call well needs them: a validator may now return the row the user reads and the result the model gets separately, a tool may word its own cancellation, and a tool may start work when its card appears rather than when it is approved. The gate is consulted before any of them. `shouldAutoAcceptToolConfirmations` is asked about the tool by name, because skipping the confirmation wait is itself an answer on the user's behalf and one tool must not be answered for. Deciding that without the name would put the exception out of reach of the only path that needs it. The gate stays inert until a chat supplies `isPlanModeActive`. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat(copilot): give a session one versioned plan document The plan the user agrees to has to survive `/clear`, so it belongs to the session rather than the conversation, and a session holds exactly one. Its id is the session's, so the primary key is the constraint — there is no second row to mint, no index to maintain and no schema change at all. Every write reads the row it is about to replace inside the transaction that replaces it. Read outside, two tabs both see version N, both stamp N+1, and the later write silently drops the earlier one's text and its snapshot; IndexedDB serialises readwrite transactions over a store, so read and write together cannot interleave. Approval takes the same route but patches only the pointer: an approval computed while another tab was revising must not carry this tab's older content back over the newer text. Approval is `approvedVersion`, a pointer at a version, never a flag. Below the current version means the newest text is a proposal the user has not agreed to; absent means nothing here was ever approved. Only exit_plan_mode can leave the pointer behind, since every write outside plan mode carries it forward — an amendment the user's posture already trusts is still the agreed plan. Declining writes nothing at all: the refused proposal stands as the newest version, with the agreed one still in history. Nor can create_artifact confer approval. It asks for no confirmation, so the model writing a plan document is not the user agreeing to one; a plan written there holds the session's slot as a draft until a decision lands on it. That is also why the approved version is exempt from pruning. A plan approved at v1 and then planned against for twenty more rounds would otherwise lose the very version that stands as agreed, and with it the card that opens it, the banner offering it back, and read_artifact at that version. It is excluded from the pruning candidates rather than added on top, so the budget is unchanged and what survives simply stops being contiguous. The write reports whether the database took it. Most callers still degrade like the reads do, but a plan cannot: returning one the database refused would let the user approve and execute against a document that disappears on reload — a refused plan write raises instead. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat(copilot): add plan mode — the posture and its two tools enter_plan_mode asks to hold work; exit_plan_mode hands over a plan and, on approval, gives the posture back to whatever preceded it. Both carry `planModeSafe`, since a posture with no exit is a trap. Only the transition the current posture allows is offered, so there is no tool for leaving a posture the chat is not in. A planning round runs from entering plan mode to the proposal the user decides on. It remembers only the write it made, because nothing it does is undone — and that write is shared between the card's confirmation hook and the tool's `fn`, so the plan is on screen while the user is deciding whether to approve it rather than after. The round is identified by an epoch bumped on *entering*, not by the conversation. A chat rotation mid-approval must still let that approval hand the posture back; a round the user has since left and re-entered must not, or approving the old plan would drop them out of a read-only posture they just chose. Saving a proposal revises the session's plan document and creates one only when there is none — both halves in a single transaction, so a second tab proposing at the same moment revises the row this one wrote rather than racing it. Persistence failures hold the posture. Approval is reported only once both the proposal and the approval pointer are durable, so a plan the database refused cannot unblock mutating tools. The failure is reported from `fn` and no earlier: the write settles while the card is still waiting to be confirmed, and clearing that card from underneath the wait would take away the only control that resolves it. An auto-accepting posture answers for the user through one predicate, asked by every path that answers: the pending-card sweep, the confirmation itself, and the decision to skip the wait at all. enter_plan_mode never qualifies: YOLO means "stop asking and run it", and a call from a tool set snapshotted before the switch must not answer that with a read-only posture — whether its card is already pending or has yet to be registered. Plan mode lives in its own controller with a narrow view of the chat it runs in: it reads that autonomy state and asks for the two changes it can cause, rather than owning any of it. Plan mode is offered only in a session chat, and a session chat is GLOBAL for its whole life. The gate reads that mode, so `changeMode` refuses to move one out of GLOBAL rather than resting the invariant on a picker being hidden. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat(copilot): surface plan mode in the chat and the artifact list Plan mode is the only posture that refuses work, so the composer says so before the user types the request it is about to turn down: the mode pill is tinted whole rather than by its icon, and the empty placeholder carries the constraint in words. Teal, not the house green — green is the transcript's success colour a few rows up, and a mode signal in it would read as "this worked" rather than "this is held". A blocked tool renders as its own lean row naming the tool, not as an error: the call did what plan mode says it should, and "why can't it edit" is answered where it is asked. A plan card names the decision — proposed, approved, or not approved — and never the button, since a Stop and a posture switch resolve it too. Its button opens the version that card proposed, so a card far up the transcript still shows the plan it put forward rather than whatever the document has become since. The artifact list and the preview header both label the plan through one badge helper, so the two cannot disagree about what counts as one: a plan the user never approved keeps the plan icon and takes the neutral badge, leaving the teal to mean exactly one thing. In the viewer, an unapproved revision says so in a bar that cannot be scrolled past, with the version the user did agree to one click away. The autonomy picker became a table with one row per posture, so adding one touches a single place instead of four parallel switch statements. A version of a plan is read against the one the user approved, not against the newest: latest is only where the model happened to stop. So the approved version is never stale — its bar is teal and points forward to the draft rather than warning about it — the version in front of it is the draft, and anything behind it is history that is neither and takes no pill at all. The list opens a plan at the approved version for the same reason, which is what lets its pill say `plan` while an unapproved draft sits at the head. One helper answers all of it, so the list and the preview header cannot drift apart on what counts as the plan. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * test(ai-evals): exercise plan mode end to end A case a unit test cannot stand in for: it starts in plan mode against the real gate and the real exit_plan_mode, and grades whether the model researches and hands over a usable plan instead of guessing at one. The checklist does not grade what the harness does for the model — exit_plan_mode writes the plan document itself, so "saves the plan as an artifact" would pass on any run where the tool is called at all. The eval store seeds artifacts with history and mirrors the store's own approval rules, so a rename cannot promote a proposal the user turned down. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(ai-evals): import the plan-mode messages from the module that owns them `PLAN_MODE_MESSAGES` moved to `planModeMessages.ts`; `planMode.ts` imports it without re-exporting. Under vitest, which runs the frontend adapters, the stale import resolved to `undefined` rather than failing to link, so `global-planmode1-hands-over-a-plan` threw on the approval message after the posture had already been dropped and the tool withdrawn. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(copilot): state plan mode's constraint in neutral text The composer's two-tone placeholder becomes a plain "Read-only" beside the autonomy picker, next to where YOLO puts its own warning, and a blocked call's row drops the mode colour. Teal is left marking what the posture is — the badge, the version bars, the pill — rather than every call it refuses. ContextTextarea goes back to main with the accent: `placeholderAccent` had no other consumer, and the aria-label existed only because the accent blanked the native placeholder. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(copilot): hold the plan header's verdict until the snapshot lands Opening a plan at the version its reader approved pins a version behind the head, and until that read resolves `shownVersion` is still the head — so the header wore the draft's badge and its orange "not approved" bar over the very case the pin exists to serve, then flipped. The header now says nothing while `restoringPin`, as the body already does. Judging `pinned` instead would print the approved signal over text that is still the draft, trading a true transient signal for a false one. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(copilot): refuse a hand-over once plan mode has ended A response can carry two exit_plan_mode calls, and the tool list they run against is snapshotted before the first one restores the posture. The second then found the tool with plan mode already over: under YOLO every confirmation is answered for the user, so it wrote its own summary and stamped the user's approval on a plan no card had shown them. Refused in `validateBeforeConfirmation` rather than in `fn`, since `onConfirmationRequested` writes the document too. The maintenance path is untouched — a plan still gets revised outside the posture with update_artifact, which is what the tool's own description already tells the model to use. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |