mirror of
https://github.com/windmill-labs/windmill.git
synced 2026-10-29 16:15:51 +00:00
394d345e0c8d9d7da87371e8da090d3becc77ce7
737
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
  hugocasaandClaude Opus 5.5
|
394d345e0c |
fix: give the agent page's API and Slack sections tabs on small screens (#11901)
* fix: give the agent page's API and Slack panel a tab on small screens Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: keep the agent pane mounted across the small-screen breakpoint Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: make the covered agent pane inert on the API tab Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: flatten the agent page's API and Slack sections into its small-screen tabs Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: fall back to the API tab when the agent has no Slack section Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> |
||
 
|
35b250e767 |
feat: let AI agents answer in Slack (#11630)
* feat: let AI agents answer in Slack Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: dedupe Slack deliveries by message and route only connected workspaces Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: key Slack conversations with the workspace secret and require signed Slack requests Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: keep a named agent on its thread after an empty question and flag a missing Slack signing secret Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: register the slack_agents token scope domain Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * feat: name the Slack bot, offer Slack only for chat agents, align the agent API tab with webhooks Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: warn when Slack can't be reached instead of showing a placeholder bot name Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * feat: chat SDK, streaming examples and a copyable integration prompt on the agent API tab Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: keep the base path in the chat SDK example, list slack_agents token scopes, bound the bot name lookup Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * feat: opt into the AI agent Slack scopes at connect, store the signing secret in instance settings, surface the integration prompt Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: dispatch Slack agents only on a verified request, warn when the connection lacks the agent scopes, register the callback redirect URLs Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: point the agent chat examples at @windmill-labs/chat-sdk Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: build the agent API examples on the configured API base URL Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: say that a channel's default agent takes its /windmill commands Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * feat: list the channels answered by an agent in the workspace Slack settings Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * refactor: keep the Slack signing secret in SLACK_SIGNING_SECRET, warn at connect when it is missing, move the integration prompt to a tab Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: wrap copyable code blocks at word boundaries Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: open the agent panel on its API tab, put the AI prompt first, list channels like the critical alert channels, cap the manifest's height Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * feat: group the agent API examples by AI prompt, TypeScript and HTTP, with React and framework-free chat SDK examples Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: have the agent integration prompt read the token from WINDMILL_TOKEN Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: trim the agent API tab intro, name the Slack app in mentions, define the SDK example's output Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * feat: share agents in selected Slack channels, run as their identity Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * feat: share agents in Slack channels with anyone or workspace members only Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * feat: set who can ask an agent in Slack and its channels from its settings Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: treat a missing Slack audience as workspace members Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * feat: list Slack channels before who can ask, name the run-as user in each choice Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: name the next deploy's run-as user in Slack choices, turn agents on by default when connecting Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * feat: answer agents on mentions only, group agent settings into sections Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * feat: show how to ask an agent in Slack first, keep agent scopes opt-in Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * feat: list agents shared in Slack in a setting card and table Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * chore: update ee-repo-ref to 10bc314c95094c0851743cbca31b50a3b37d42fd This commit updates the EE repository reference after PR #865 was merged in windmill-ee-private. Previous ee-repo-ref: 47d8fbac461171bd93d888b49cc1c7730a2b3f96 New ee-repo-ref: 10bc314c95094c0851743cbca31b50a3b37d42fd Automated by sync-ee-ref workflow. --------- Co-authored-by: Claude Opus 5.5 <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> |
||
|
hugocasaandClaude Opus 5.5
|
0b980546b1 |
feat: agents run on behalf of their deployer (#11430)
* Revert "docs: agent editor comments describe runs as the caller" This reverts commit |
||
|
08319a004c |
feat: allow early return on steps inside branches (#11733)
* feat: allow early return on steps inside a branch to one The "Early return for sync webhooks" flow setting could only point at a top-level step. It can now also point at a step inside a branch to one, in any branch including the default one, and through nested branches to one. When the branch is taken, the sync call returns that step's result as soon as it completes while the rest of the flow keeps running. When it is not taken, the call returns the flow's final result. The sync wait loops already resolved nested steps and fell back to the final result, so they are unchanged. Worker: check_if_early_return_or_last_in_early_return_parent now accepts a job whose step id matches early_return at any depth, as long as every job between it and the root is a branch to one. A loop iteration or a subflow reusing the same id does not take the stream. The existing case, the last step of a branch to one that is itself the early-return step, is unchanged. Frontend: the native <select> is replaced by EarlyReturnPicker, a searchable list in flow order where each nested step shows its path (e.g. `a › Branch 1`). Branch to all and loops inside a branch to one, and the steps inside them, are not offered: the results lookup cannot resolve them when nested. Picking a nested step shows a hint that it only runs if its branch is taken. * fix: stream the last step when the early return branch is not taken Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: keep the stream off the last step once the early return step ran Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * refactor: list early return steps by id only Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * refactor: use the standard select for the early return step Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: match the early return step by its flow scope, not its path Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: keep branch to all and loop steps off the early return stream Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * feat: allow early return on steps inside a branch to all Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: resolve a nested branch to all or loop as the early return step Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * test: pin which step streams for an early return inside branches Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: return an empty nested loop or branch to all as its early return list Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * revert: read an empty container's early return status from the root An empty nested loop or branch to all falls back to the flow's final result again, so an empty top-level loop that failed keeps its real status. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> --------- Co-authored-by: stefan <stefan.radovanovic@ip-paris.fr> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> |
||
|
hugocasa
|
9249aa02d5 |
feat: let a linked agent step override model and reasoning effort (#11685)
* feat: let a linked agent step override the model and reasoning effort * fix: only honor model and effort overrides the step authored * fix: treat an authored empty effort override as no effort * fix: keep a static empty override from reading caller arguments or the saved effort * fix: keep step tests and multi-agent chats from inventing an effort override * fix: keep the saved model fallback when linked steps differ only on effort * fix: show the agent override hint on the input's own help row * fix: align the override hint spacing between static and expression inputs * fix: keep the override rows a dropdown whatever the model is set to * fix: give the override hint the 4px form gap and drop its trailing period * fix: space the override hint with a margin so it is not clipped * test: pin the saved model fallback when linked steps differ only on effort * fix: keep the saved model fallback across agents that saved different efforts |
||
 GuilhemandClaude Opus 5.5
|
3857ce3693 |
feat: shrink the page header in stages and fold page actions by measured width (#11732)
* feat: shrink the page header in stages and let the bar fold page actions Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: fold the runs header through the bar and wrap its filter to a second row Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: truncate the run id and fold run actions into one menu on a narrow bar Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: keep an item's summary and drop its path on a narrow page header Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: show the full summary and path read-only when they cannot be edited Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: keep secondary detail actions in menus and the star beside the summary Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: show secondary detail actions as buttons while the bar has room for them Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: fold embedded detail headers, refresh stale fold widths, keep the path beside a summary placeholder Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: drop the duplicate path on read-only detail pages and re-measure folded actions when they change Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: re-measure the runs header when its hidden counters change width Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * revert: drop the runs header's counter version, the badges do not change its width Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: stop the header re-measuring its actions every frame while idle Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com> |
||
|
Ruben FiszelandClaude Opus 5.5
|
0758bdc46b |
refactor: remove the frontend build-time base url (#11674)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> |
||
|
GuilhemandClaude Opus 5
|
338c4e921d |
feat: operators get AI sessions (#11443)
* feat: operators get AI sessions `prefersSessionHandoff` stops taking the role, so every AI entry point an operator can reach — Ask AI, the home composer, Open in session, the flow editor's hand-off, the raw-app template picker — opens a session for them like anyone else. The sessions page drops its refusal, and the layout stops keeping them on the docked chat. In the sidebar they now get the Workspace/AI Sessions switch and session mode, and lose the Ask AI entry that stood in for them, since the pane behind it is no longer the one they are sent to. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: operators get run-oriented examples in the home composer Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * docs: note builder_flows in the operator composer examples comment Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: gate session tools and editors on operator rights Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: keep operator AI sessions behind a browser opt-in Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: keep flows refused in the prompt for operators without builder rights Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * refactor: decide every frontend edit check through one editRights module Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: let admins who are also operators open session editors, as drafts.rs does Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: mirror drafts.rs exactly in roleCanDraft and correct the edit-rights docs Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
hugocasaandClaude Opus 5.5
|
c52ba2fd30 |
feat: agent nits: multi-select, memory-off notice, agent step label (#11622)
* feat: agent nits: multi-select, memory-off notice, agent step label Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: say saved, not renamed, on a new agent's first deploy Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: run the agent page form on cmd+enter, select agents by keyboard Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * test: pin the agent step label Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> |
||
|
Ruben FiszelandClaude Opus 5.5
|
a85d0b76cc |
feat: add Pi template to the AI sandbox picker (#11619)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> |
||
|
|
41162b270c |
feat: one page header across routes, and one sidebar for operators and developers (#11310)
* wip: one sidebar for operators and developers, with a detach mode Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: snap-detach the operator sidebar, and join the handle to the picker Dragging the sidebar into the left snap zone detaches it for an operator, who has no icon-only rail to collapse into; the width the drag started from is what gets persisted, so re-attaching restores the sidebar's size. The handle's colour pastille slides onto the card's workspace pastille as the card opens, alongside the workspace name, and both buttons that attach or detach the sidebar carry a tooltip. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: attach and detach from the sidebar header, open the card under its handle The button that swaps the sidebar between docked and detached moves to the header, next to the workspace name, so it stays in one place across the swap; the icon-only rail has no room for it and keeps the footer copy. The handle itself becomes a component of its own, carries the workspace icon its picker uses, and opens the card rather than docking — docking is the header button's job. The card is placed under the handle measured at open time instead of at a fixed offset. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: one page header across routes, with the workspace picker in it Every route now wears the same bar above its content: the sidebar toggle, the workspace picker, and whatever the route registers — an item's breadcrumb and summary, or a section name — with the route's own buttons at the far end. The bar's row is level with the sidebar's own top row, which the workspace ⇄ sessions switch now owns, so the two read as one band. A route fills the bar by mounting PageHeaderContent, which registers a getter rather than a snapshot: the layout calls it while rendering, so a path or summary the page edits reaches the bar without an effect keeping two copies in step. Nothing registers yet; the routes follow. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: detail pages fill the page header, which now spans above the sidebar A script or flow detail page no longer draws a bar of its own: its path becomes the header's breadcrumb, its summary keeps the rename-and-labels popover next to it, and its badges, menu and buttons ride along as the header's actions. The band runs the full width with the sidebar hanging below it. A snippet rendered in the header initialises its components under the header, not under the page that wrote it, so anything the page's tree provides is lost — the detail page's buttons read TriggerContext and threw. ContextBridge carries the entries a registration names back around the render. The floating handle is gone with it: the header's own toggle is the one control, hovering it peeks the sidebar in and clicking pins it. The workspace picker in the header is a denser variant of the sidebar's, and the card's outside-click catcher starts at the card rather than the viewport top, so it no longer swallows clicks meant for the toggle. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: one breadcrumb in the page header, and the banners that used to sit under it The header band now carries a single breadcrumb — workspace, fork, then the item's path — built from one component, so every segment has the same size, weight and icon size, and slashes separate them. The item's kind icon rides with its summary, after a dot, rather than with the path. The fork and drafts banners move into the band as badges plus a subtle Review button: ahead/behind counts instead of a bare change total, and the per-kind breakdown stays on the compare page. The home hero's CLI, MCP and Hub links move into an ellipsis menu, alongside the toggle that pins the session chat to the page. The flow editor and the app viewer hand their bar to the band too, and a route can drop the band's bottom border when its own content draws one — the border stays in the box with a transparent colour, so contents never shift a pixel across a navigation. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: the runs page's own header row moves into the page header The title, the queue counters, the kind and status filters, the schedule and future-job switches, the search field and the timeframe picker all ride the band now, and the page starts at the chart — one row of vertical space back. The two toggle groups appear from 1500px: below that the bar cannot hold them next to the search field, and the same filtering is reachable from the search field's own kind and status chips. The search field is what gives width back, and the breadcrumb yields last, so a page's own name stays whole while the workspace and fork names truncate. The queue counters keep their icon form at every width, since the bar has room for two counters and never for the two labelled blocks. Their counts hang outside their icons, so the group carries the margins that keep a badge off its neighbour, and the suspended counter answers a hover at zero rather than sitting there as a dead icon. The band is 44px rather than 40, and a page can hand the breadcrumb a hint — the documentation tooltip that belongs with the page's name rather than with its buttons. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: no border under the page header, and the peek card closes on the way out The band draws no bottom edge and no page asks it to, so `barBorder` goes with it — a page that needs the separation draws its own line, as the detail pages now do above the run form and the tab strip. With nothing toggling a border, the band is a flat 44px on every route. The detached sidebar's card is a hover affordance, so it leaves when the pointer does: off the handle and off the card both, on the same debounce a hover menu uses, because the two do not touch and the leave of one fires before the enter of the other. The mobile drawer is opened deliberately and still stays until it is dismissed. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: one band, one workspace part, and the pickers that hang off it The band is a single element whose placement is styled rather than branched: a page registering a side-panel inset used to swap it between two `{#if}` arms, destroying the header and rebuilding it a frame later — a blink of no header with the content sliding up into the gap. The workspace and its fork become one breadcrumb part: the name leads home (inert in a session, where the workspace is the session's scope rather than a place to go) and one chevron opens the picker, which already expands a family to reach its forks. The picker's panel hangs from the part's left edge, opens with the active family expanded, and closes before the switch runs — the switch renames the workspace, which moves the trigger the panel hangs from, and floating-ui was chasing it across the viewport. The item's path is a plain label rather than a row of pickers: same separators and weight, and a click copies it with its own confirmation. The kind icon goes, the page below says what kind it is. Around them: a "prod" badge for a root workspace, the session's acting fork beside the session's name rather than in the breadcrumb, the edit-in-dev affordance as a button in the header, the drafts count as a sentence, and a fade when the workspace changes under a page that stays mounted. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: keep a page's own controls when the workspace navigation is hidden `menuHidden` means "this is an embed, don't show the workspace's navigation" — but Runs, the detail pages and the flow editor now keep their controls in the band and nowhere else, so hiding it took the filters, the timeframe and the page's buttons with it. A session's preview iframe opens exactly that way. The band now renders for an embed too, without its breadcrumb, so the page keeps its controls and the embed still gains no way to navigate the host's workspace. The flow editor's autosave indicator moves into the header's actions with the rest of its bar: saving state, the autosave toggle and other people's drafts were left behind in the branch the full-page editor no longer takes. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: the header carries what it promised, and nothing it does not The registry's item described pickers the breadcrumb no longer has: nine fields — the editability flags, the change callbacks, the navigation hook — that nothing read, while the flow editor passed eight of them and quietly lost its inline rename. The type now says what the bar does: a path, and either a summary or the page's own control for one. The flow editor renames through that control, as the detail page already did. `barPlacement` went the same way: no page registered 'inline', so the layout's guard never fired. The sessions page registered an inset of 0 whenever its preview pane never mounted — the "no session selected" and opted-out arms — and 0 reads as "a page owns the right edge", floating the band over the page's first line. It registers an inset only once the pane has a width. The band's sidebar handle docks through `setDetached` like the card's own toggle, so docking still suppresses the rail's entry animation and closes the card; a summary without its breadcrumb no longer leads with a stray separator in an embed; a page's hint sits with its name whichever way the page named itself; and `Menu.closeInstantly` and `EditorHeader`'s `inline`/`hideSummary` go, their callers having been replaced. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: a deployed app gets the viewport, and the header waits in the corner An app is the page, not a page inside Windmill's chrome, so the band leaves the flow and floats above the viewport until the handle in the app's corner calls it down. The handle is faded until the pointer nears it, hovering peeks the band over the app, and a click pins it — one flag for every deployed app, in this browser, because someone who wants the breadcrumb over a dashboard wants it over the next one too. Peeking overlays rather than pushes: the app is an iframe, and taking 44px off its box on every pass of the pointer would re-lay-out whatever it draws. Pinned, the band is in flow like any other route's and the app starts under it. The sidebar is untouched — docked stays docked, detached stays a peek — since the pin is about the header alone. A page says it wants the viewport with one registry field, so the layout keeps asking the registry rather than knowing about apps. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: one corner control for the chrome an app page hides Hovering the button overlaid on a deployed app brings back whatever that page is hiding: the band always, and the sidebar card too when the sidebar is detached. The card hangs from this button rather than from the header's own handle, which is off-screen with the band. The handle in the band no longer attaches the sidebar — it reveals it and nothing more. A control that hides the thing it sits on cannot also be the way to bring it back, so attaching stays with the toggle in the sidebar's own footer, where detaching happens. The band's two buttons go in the order the things nest: the sidebar's handle, then the header's pin. The overlaid button stands where the pin will be once pinned — same 28px box, and a slot further along when the sidebar's handle takes the first — so pinning moves nothing. It wears the panel pair the sidebar's toggle uses: solid while the band is away, dashed while it is shown. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: the access point sits in the corner, and the header covers it The button is in the page's top-left corner — 8px in, or just past the rail when the sidebar is docked — rather than shifted along to where the header's pin would land. Hovering brings the header down over it: an access point that the regular view covers once it arrives, so there is nothing to line up and nothing to move. Because the header lands on the button, the pointer leaves it the instant it arrives. The band holds both itself and the sidebar card open while the pointer is anywhere on it, or the card would retract under a header that stayed. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: on an app page the floating button owns both, and they move together The band drops its sidebar handle there: the floating control in the corner reveals the sidebar along with the band, and two handles for one gesture is one too many. The two are then linked — the pointer resting on either holds both, and leaving either collapses both — so they behave as the single thing the one button revealed. The card meets the header in a line: it hangs from the header's bottom edge rather than from whichever control opened it, which had it starting a couple of pixels high. With no header to sit under — an embed — it still hangs from the control that reached for it. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: the revealed sidebar is a panel under the header, not a card over the page No rounded corners, no inset from the window: it starts at the header's bottom edge and runs to the left and bottom edges, with one border down its right side and a shadow to lift it off the page. A card floating in from nowhere read as a separate thing; this reads as the sidebar the header sits on top of. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: only an app page gets the flush panel; elsewhere the card still floats Over an app the sidebar reads as the panel the header sits on — flush left and bottom, square, one border down its right side. On every other page it is what it was: a card inset from the window with rounded corners and a border all round. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: an app opens with the header already away Opening an app showed the header and then slid it out: the page claimed the viewport on mount, and the band animated from where it had been rendering. The slide belongs to the hover, not to arriving, so it is armed a frame after the claim — the first hide is a cut. The claim moves to the route, which mounts before the viewer waiting on the workspace to resolve. A cold load now paints the band away from the first frame instead of showing a header that leaves again. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: an app's Edit sits with its name At the far end of the bar it was a journey across the window from the thing it edits — on a page whose header is only there while the pointer holds it, a journey with a time limit. It goes beside the app's path instead. The slot it uses is the one Runs' documentation tooltip already had, renamed from `hint` to `afterName` and documented for both: what belongs to the page's name, whether that is a mark or a control that acts on what the name points at. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: developers collapse, operators detach One control each, rather than three for everyone. A developer switches the sidebar between icon-only and full width and never detaches it; an operator, who has no icon-only rail to collapse to, hides it instead — by the button in its footer or by the same drag that collapses it for a developer. The way back survives either way: a detached sidebar keeps its "Show sidebar" button for anyone, so a developer who detached before this was so is not left with a sidebar they cannot return. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: a page is as tall as it is given, not as tall as the screen The band takes 44px off the top of every route, so a page sized to the viewport overflows by exactly that much: h-screen inside a column that already lost the band's height. These pages measure the box they are in instead. CenteredModal gains `inPage` for the five pages that render it inside the layout rather than over it — the same modal, centred in what is left of the page rather than in the window. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: the band narrows, wraps, and draws its own edge The band publishes its width, so the pages filling it can answer it. Two tiers: under 1000px the breadcrumb keeps the workspace disc and its env badge and drops the name, and under PHONE_BAR (800) the trail stops yielding at 10rem — that floor is what makes the bar wrap, so a page's buttons take the line below rather than squeezing the name to an ellipsis. The band also draws the line between itself and the page, and only while something is scrolled under it. One capture listener on the window sees every scroller, since most pages scroll inside their own box and never move the layout's. An inset shadow, not a border: a border would move every page down a pixel the moment the line appeared. DetailPageLayout and InWorkspaceAppViewer drew that first line themselves; two lines would stack, so theirs go. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: the script, app, runs and audit pages fill the band Each of these kept a header of its own above the page. They register with the band instead: the breadcrumb names what the page is on, and the page's buttons sit at the far end of the one row. They answer the band's width rather than the window's, since the buttons share that row with the breadcrumb. The script editor folds Diff and Settings into the menu it already shows when compact, the app editor drops the panel buttons, Awareness and Export and goes icon-only on Deploy and the mode switch, the flow editor menus its diff, and runs hides its counters. Runs also declares `actionsFlexible`: its search is the one control here that can give width back, and a box that refused to shrink would make it take its maximum instead. The run detail page loses its ActionRow, and the audit log its big title — the breadcrumb says where you are, and the page keeps its tooltip. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: one sidebar column, shown as a rail or as a panel The rail and the drawer held two copies of the same column — the same sections, in the same order, in two pieces of markup, so a change to one silently skipped the other. Both now render `sidebarColumn`, and the shells keep only what differs: the rail its width, resize handle and collapse button, the panel its outside-click catcher and its transform. A phone gets that panel rather than a modal. There was a width test deciding between two presentations; the card under the band's handle works at any width, so it goes. What is left of the width test is the choice between a rail and a panel, plus a `(hover: hover)` test — the old `innerWidth >= 768` was standing in for "has a pointer", which is what decides whether the panel closes when the pointer leaves it. With the card everywhere, the backdrop, the `aria-modal` (nothing dims and nothing traps focus) and the drawer's close cross all go: the band's handle opens the panel and a tap outside closes it. The burger row goes with them. Entering panel width now collapses the rail transiently, so coming back to a wide window does not land on a rail the page was resized away from. The band also stops easing its right edge. That transition belongs to the rail, which moves the band's left; its right follows a splitter the page owns, and easing it left the band lying over a session's side panel — a blank strip across that panel's own header for 200ms after every drag. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: the home composer flies to the menu that restores it Removing the composer puts it away rather than throws it out, and the way back is one item in the band's menu — easy to miss on a page the composer has just vacated. A mark flies from the control that removed it to that menu, and the menu answers with one ring as the mark dissolves, so where it went is witnessed rather than explained. The cross and the menu item now read "Remove session chat", one name for one thing whichever a reader reaches for. Reduced motion skips both. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * revert: keep operators out of AI sessions That change is its own PR; this one is about the navigation. `prefersSessionHandoff` takes the role again and every entry point passes it, the sessions page refuses an operator arriving by URL, and the layout leaves them on the docked chat. Two clauses are new rather than restored, because the nav rework is what gives an operator a sidebar at all: the Workspace/AI Sessions switch and session mode are both hidden from them, since the page behind them declines. `askAiShown` gains the operator case for the same reason — they keep the docked pane whatever the beta gate says, so the sidebar has to offer a way into it. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: keep every control the band took over reachable Four controls the page-header conversion could put out of reach, each found by a review pass and reproduced in the browser: - The flow editor's preview buttons. `topbarWidth` is bound only in the branch a full-page editor does not render, so `compactTopbar` was permanently false there; below the 800px phone bar `phoneTopbar` dropped the buttons inline and the graph overlay never took them, leaving FlowPreviewButtons mounted nowhere and step testing, whole-flow testing and "Test flow & record" no-ops through optional calls. It now measures the header the way ScriptBuilder and AppEditorHeader do, and the inline and overlay conditions are exact complements. - The pin on a deployed app. The band lands over the corner handle that called it down, so the pointer that arrives there hits the band. The band carries the pin itself now, on the same pixels, in place of the sidebar handle it hides. - The detached sidebar's own menus. They portal to `body`, so moving into one left the card and closed it 200ms later, sliding it out from under the menu being read. The card holds while the pointer is in a menu and takes the close from the leave of that. - A deployed app's name for a read-only viewer. The breadcrumb was registered behind write access, so without it the trail fell back to the section name "Apps"; and the Edit button rode `afterName`, which the band drops for an embed along with the trail. The name is registered for whoever opened the app, and what sits after it survives `navHidden` as the page's own control. navPageFor gets a table test: it answers both the sidebar links and the breadcrumb label, and its longest-prefix and first-segment rules are what made the "Apps" fallback look like a correct answer. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs: drop what the operator revert and this refactor left behind The operator tour's last step named AI sessions among the pages the sidebar opens, which operators have no way to reach. The nav handle's comment named the AI chat layout's menu button as a second way into the card; that button and its prop are gone, and the corner control on a page that hides the band is what stands in for the handle now. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * style: prettier the three files this branch left unformatted `prettier --check` fails on all three at the previous head. Whitespace only, apart from one Button tag that fits on a line. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: give a page without the band's chrome what the band was carrying Second review round, three reachable defects plus one decision: - A public raw app was 150px tall. The bundle's iframe is `h-full`, and dropping its `min-h-screen` left nothing to resolve against on the share, embed and custom-path routes — none of which has a layout above them. Measured: 150px in a 900px viewport. The floor goes back for those routes and stays off inside the workspace, where the layout hands a height down and `min-h-screen` would overflow it by the band. - The band retracted out from under its own open menu on a deployed app. Last round taught the sidebar card to hold while the pointer is in a portaled menu and left its sibling alone, but the band opens menus too — the workspace disc, the fork family, every path level. One helper now, used by both closers. - `bandPeeked` outlived the page it belonged to. Clicking a control in the band navigates away, and the next page may not be one the band floats over, so the hover handlers that would clear it are gone — leaving the band already down when a full-bleed page came back. `beforeNavigate` clears it. - An embed has no band to carry a control, so the app's Edit floats over the canvas there, where it sat before the band existed. Registering it as the band's `afterName` under `navHidden` instead would have cost the app 44px of the iframe for a row holding one button. Reading "is the workspace navigation hidden" now has one definition (`isMenuHidden`), since a page has to answer it too: the flag is sticky inside an iframe, so the raw query param is the wrong thing to read. Also inset the band by the rail only when there is a rail — an embed reserved its width and crowded its own controls into a compact layout early. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: a path that names nothing yet has no breadcrumb leaf `splitItemPath` documents that it gives up on a path still being typed, but `f/demo/` is three parts and returned a leaf with an empty name — a blank segment in the trail. The editor binds the path live, so the breadcrumb sees every keystroke. Tested here rather than `navPageFor`'s sort: no two `NAV_PAGES` paths nest, so the longest-prefix case that test claimed to pin has only one candidate. Renamed it to what it actually checks. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * style: prettier the four remaining files this branch left unformatted All four fail `prettier --check` on the branch and pass at the merge base. `raw_apps/utils.ts` also fails but already did at the base, so it is left alone. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: the raw app editor fills the band like the other three It was the one editor route still rendering its own bar under the band, so it showed two header rows and the band named the app "Apps raw" — `navPageFor`'s first-segment fallback, because nothing on the route registered an item. Its top bar now becomes the page header the way the script, flow and low-code app editors' do: the path and summary are the breadcrumb's, the file-sidebar toggle, autosave cloud, presence mark, menu, Diff, Jobs, export, AI and Deploy are the header's actions, and the path is a plain label since this editor renames from its deploy drawer. The actions render under the band, outside the subtree `RawAppEditor` declares the operating workspace in, and `OpenInSessionButton` asks that context who the acting user is — so the workspace travels with them. `operatingWorkspaceContextEntry()` hands it over without exporting the key. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: keep the folders a half-typed path already names The trailing-separator guard gave up on `f/demo/` entirely, which dropped the whole trail to the kind chip and brought `f/demo` back on the first character of the name. The levels the path does name come back now; only the leaf waits. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: pin holds the first slot, and the nav table says what it shares Pinning brings the sidebar's handle back, and from behind it the unpin button sat one slot over from where the pin had just been clicked — so a second click in the same place opened the sidebar instead. An operator is detached by default, which makes that the default state of the feature rather than an edge case. `NAV_PAGES` claimed one table stops a page being called one thing in the nav and another in the header, which holds for the seven entries the sidebar's main group is built from and not for the four grouped separately — `/routes` was "Triggers" in the header and "HTTP" in the nav. The comment now says which is which, and that entry is named for the page rather than its group. `sidebarPageAllowed` gets a table test: it decides what an operator sees, and its `=== true` is what a future `Option`-style default would silently change. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: the band names the path Deploy will create, and keeps the summary readable A brand-new app is parked at a `draft_<uuid>` placeholder, and the band named that rather than the path its Deploy will create — `newEditedPath` refuses the placeholder for exactly this reason, and the editor's own bar read it. Both app editors register it now. Verified: the trail reads `u/admin/gentle_app` while the URL is still the draft. The band's summary gets the `title` the editor's own bar had, so one the bar truncates can still be read on hover. The wrapper that hid AppExportButton on a phone never hid anything: the component draws no trigger — it is the export drawer, which portals to `body` — so the class toggle was inert in both states and the comment read as if something moved. Dropped from both app editors. Two comments corrected: `NAV_PAGES`' `icon` does reach the breadcrumb for all eleven entries (`NavBreadcrumb` renders `routePage.icon`), so only `operatorKey` goes unread for the four the sidebar groups separately; and the raw editor's phone note described folding that does not happen — the file-sidebar toggle stays in the band, Jobs and the export are already in the menu, and AI is what stands down. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: an item the picker files under no kind keeps its name in the band The merge brought main's agents detail page, which reaches DetailPageHeader with no `errorHandlerKind` — and the band's item required one. `WorkspaceItemKindIcon` falls through to the script glyph for anything it does not know, so defaulting would have labelled an agent a script; the kind is optional now and the trail names the item without claiming it is one of the three. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: the file sidebar's toggle leaves the band The band is the workspace's header, not the raw editor's layout controls. ⌘B still folds the file sidebar, and the editor's own bar keeps the button where it draws one — a session pane. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: the scroll edge stops measuring, and the comments describe the code The band's edge listener captures every scroller in the app — Monaco, the audit table, a run's logs — about once a frame while any of them moves, and it took two `getBoundingClientRect()` reads before looking at `scrollTop`. It now answers from `scrollTop` first and only measures when that answer differs from what the band draws, which is the transition rather than the steady state. The rect check still authorizes every change, so a scroller that does not start at the content's top still cannot claim the edge. The two agent routes main just added size themselves to the viewport, which is 44px more than the box the band leaves them. They measure 856 either way — the layout's `min-h-0` column clamps them — so this is the sweep the rest of the branch already had, making the height follow the box rather than depend on a parent's clamp. Six comments described the conversion rather than the code, which AGENTS.md calls out by name. Each now states the constraint a reader could break: why the summary keeps its popover here, why the audit hint is `afterName` and not an action, why a run registers a `section` rather than an item, why the Runs filters are `actionsFlexible`. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * Revert "fix: the file sidebar's toggle leaves the band" This reverts commit |
||
|
GuilhemandClaude Opus 5.5
|
b2ef8478c6 |
vertically center flow step header actions (#11601)
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com> |
||
|
hugocasaandClaude Opus 5.5
|
c75113a235 |
capitalize AI Decision labels to match AI Agent (#11583)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> |
||
|
Ruben FiszelandClaude Opus 5.5
|
9d55a9381d |
add customUi prop to hide save to workspace on flow steps (#11566)
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com> |
||
|
Ruben Fiszel
|
8c74f75de3 |
fix: simplify the run page AI card to a prompt and a chat button (#11561)
* fix: drop the add a prompt for AI button from the run page * fix: open the run page in the session preview from a plain chat button * fix: rename additional prompt for AI to AI prompt * fix: put the run page chat button beside advanced when no AI prompt |
||
|
|
3cba8a7e4c |
feat: add an AI decision flow step (TypeSafe's Jev, Cloudflare's Clef) (#11474)
* feat: add an AI decision flow step on TypeSafe's Jev An `aidecision` step asks TypeSafe's Jev typed questions (choice, score, yes/no) about a state and returns calibrated answers. It is a plain step: a flow routes on the answers with a Branch to one after it, which the editor builds and keeps in step with the questions. It can also be an AI agent tool, where the calling model supplies the state. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: read single-quoted branch conditions and flag branches on a removed question Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: route decision tools by their tag and keep structured questions as JSON The questions editor writes only on an edit, so opening a step leaves its value untouched, and questions the cards cannot hold stay in a JSON editor. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: re-pick the questions editor mode when the value is replaced from elsewhere Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: keep the questions JSON editor's text in step with the value Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix: recreate the questions JSON editor when the value is replaced from elsewhere Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * feat: run AI decisions on Cloudflare's Clef as well as TypeSafe's Jev A `cloudflare` provider kind reads the hub's `cloudflare` resource (token and account id), calls the model's Workers AI URL, and unwraps Workers AI's `result` envelope. Like TypeSafe, it serves decision steps only. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * feat: branch from each AI decision question, and align its editor with the brand guidelines "Branch on this question" now sits in each choice question's card, which also links to the Branch to one that routes it and flags options it misses or branches it no longer matches. Labels, errors, cards and icons follow the form guidelines and the Branch to one's cards. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * feat: branch on an AI decision from its graph node A decision with a choice question nothing branches on yet shows a button beside its node that inserts the Branch to one, as its question card does. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * fix: resolve Cloudflare resources in the AI proxy, and use the design-system button on the graph Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * fix: put the decision_inputs test's doc comment back on it Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * chore: regenerate the flow schema in prompts and the chat after the AiDecision description edit Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * feat: switch AI decision questions between the form and JSON JSON is always available; the form is offered whenever its cards can hold the questions, and a value they cannot hold still opens in JSON. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * fix: show step ids as badges and use the JSON toggle of the run forms Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * fix: quote option names in the AI decision branch alerts Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * feat: remove stale branches from the AI decision question card An empty branch is removed directly; one with steps asks first, naming how many steps go with it. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * feat: branch on yes/no AI decision questions, and test a decision step from the AI chat A yes/no question branches like a choice: "Branch on this question" writes one "yes" branch on `noul >= 0.5`, with the default as "no", and the helper still reads it once the threshold is tuned. The AI chat's step test runs a decision alone as a one-step preview, taking the state the chat passes. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * fix: record removing several stale branches as one undo step Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * fix: name the typesafe resource type TypeSafe Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * fix: store TypeSafe as the typesafe resource type's display name, not in the word map Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * fix: leave the typesafe and cloudflare resource types to the hub instead of seeding them Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * docs(AGENTS): resource types ship through the hub, not migrations Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * fix: show the questions JSON format beside the JSON toggle Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * fix: give the questions JSON format in the field's description Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * fix: style the decision's graph branch button like Add branch Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * fix: show the questions JSON format only in JSON mode Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * feat: show each question's branching under the questions JSON editor Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * fix: drop the redundant In JSON prefix from the questions format Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * fix: shorten the decision state's description and show an example as its placeholder Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * fix: escape a schema placeholder before the JSON editor renders it as HTML Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * fix: give the decision state's placeholder a text and an object example Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL * chore: update ee-repo-ref to 154cbccb6b8d47bf662b5507882917b5933d5a45 This commit updates the EE repository reference after PR #839 was merged in windmill-ee-private. Previous ee-repo-ref: eaf38c1388f3cc5bece163404fffa959ede8cae0 New ee-repo-ref: 154cbccb6b8d47bf662b5507882917b5933d5a45 Automated by sync-ee-ref workflow. --------- Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com> |
||
|
|
e7fc1b2e2e |
feat: restricted job tokens per script and flow (#11484)
* feat: restricted job tokens (job_token_scopes on scripts and flows) Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: admit flow-run reads, skip dedicated workers, gate on worker version Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: keep restricted jobs off flow runners, preserve scopes on rename and promotion Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: keep restricted jobs off every dedicated handoff, confine progress flow id Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: exclude restricted runnables from dedicated worker startup Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: gate restrictions on the release after 1.821.0 Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: store per-job scopes on job_perms instead of v2_job, pin inline runs to the checked version Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: step-level job_token_scopes for flow steps and agent tools Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL * fix: fail closed on perms read errors, refuse restricted queue imports, gate step scopes in previews Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL * fix: carry a job's scopes on its completion so a re-run keeps the caller's cap Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL * fix: carry a zombie job's scopes into its completion Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL * fix: leave a zombie for the next sweep when its scopes cannot be read Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL * docs: correct the QueuedJobV2 completion comment Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL * fix: validate step scopes in batch flows, fail closed on unvalidated step scopes Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL * fix: refuse flows with step or tool restrictions at push while an older worker is live Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL * fix: apply the step-scope worker gate to flow restarts Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL * fix: list the job token toggle with the other step and flow settings Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL * chore: pin the EE companion merged with EE main Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL * perf: skip scope lookups for unrestricted jobs; list job token setting last Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL * style: rustfmt scopes tests Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL * fix: confine restricted job tokens to their own run lineage; drop remaining extra lookups Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL * chore: update ee-repo-ref to 259ad3bfeef5285ba80eedc86309b11dca001220 This commit updates the EE repository reference after PR #843 was merged in windmill-ee-private. Previous ee-repo-ref: 2b77c0225dca441235daf7bf0a06ba968df0c927 New ee-repo-ref: 259ad3bfeef5285ba80eedc86309b11dca001220 Automated by sync-ee-ref workflow. --------- Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com> |
||
|
Ruben Fiszel
|
e952298f84 |
feat: replace the AI input filling toggle with an additional prompt (#11501)
* feat: replace the AI input filling toggle with an additional prompt for AI * feat: pass the additional prompt for AI to MCP clients and shared AI guidance * feat: shrink the run page AI card to a button and a collapsed prompt * fix: offer writers a way to add a prompt for AI from the run page |
||
|
Ruben FiszelandClaude Opus 5.5
|
f2393e5b24 |
keep flow priority input inline with its toggle on wide screens (#11500)
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com> |
||
|
GuilhemandClaude Opus 5.5
|
88d0cd00e5 |
fix: restore the save to workspace button on inline flow steps (#11469)
* fix: restore the save to workspace button on inline flow steps Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: make the inline step save to workspace button icon only Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com> |
||
|
hugocasaandClaude Opus 5.5
|
8c7dcbbda5 |
feat: agents as a standalone kind with home listing, detail and editor pages (#11332)
* feat: list agents on the home page and create them from the new menu Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: keep runnables out of the agents view and anchor a new agent once saved Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: autosave a new agent's first edit and list agents past one page Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: add agent detail and editor pages Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: explain when an agent cannot be run and drop the broken agent move Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: keep query params and a unique path when creating an agent Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: lead the home list with agents and keep rows while the agent view loads Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: treat a loading agent as undeployed when leaving the editor page Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: agent detail page config panel, run page on form runs, chat badge Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: agent configuration modal and draft paths like other new items Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: keep a new agent draft-free until the first input, land agents with runnables Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: place AI agent after apps in the new menu and describe chat and flow use Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: new agents start with managed memory, editor form says how to turn it off Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: clearer managed memory hint in the agent editor form Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * refactor: share the agent editor's pane notice as a PaneNotice component Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: name a new agent after a path no resource holds Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: tell repeated tool names apart and hide permissions on draft-only agents Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: agent configuration beside the model in the chat composer and above the form Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: center the agent run form, configuration beside its Run button Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: icon-only agent configuration button beside Run Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: agents run on behalf of their deployer through a run-by-path endpoint Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: agents keep their run-as identity in their value, preserved by the CLI The identity an agent runs as lives in `value.on_behalf_of` instead of a column. Every write of an agent resolves it server-side as a flow's is: the writer's own, unless an admin or wm_deployers member asks to keep it, and a folder default on create. Retyping a resource into an agent resolves its value the same way. Export leaves it out; the run endpoint reads it and drops it from the step's inputs. The CLI follows the app model: a pushed agent never takes its identity from the tracked file, an unchanged agent compares equal to the deployed one, and an admin or deployer push claims the deployed identity back. The owner-change pre-check lists agents. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: deploying an agent draft from review keeps its deployed identity As for an app: an agent draft carries no identity, so the review page claims the deployed one back, which the backend honours for an admin or wm_deployers member. The draft diff leaves the deployed identity out, since a draft never holds one. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: deploying an agent to another workspace offers the run-as choice An agent deployed to prod/staging, merged from a fork or promoted with `wmill workspace merge` gets the same identity choice as a flow or an app: the target's current one, the deployer, or a picked user, sent as a principal the way a trigger's is. The source workspace's principal is never copied, and a difference in identity alone is not a change in the workspace compare or its diff. The frontend consumes the published windmill-utils-internal, so its deploy provider carries the same rewrite until that version ships. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: agent identity on fork, agent-scoped job reads, and the run license gate A fork re-points an agent's identity at its creator when they may not preserve someone else's, and at the creator when it names nobody in the fork, as it does an app's. A token scoped to `jobs:run:agents:<path>` reads back the runs it starts, chat turns included. The agent run endpoint checks the enterprise license like every other run entrypoint. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: a CLI push decides agent identity handling by the tracked file's type An agent retyped into another resource by a push kept neither its value's `on_behalf_of` as the file stated it nor clear of the old agent's identity. The file's type now decides, and only a deployed agent's identity is claimed back. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * refactor: agents run as the caller, with a note on what a shared one needs Drops the agent run-as identity: its storage in the agent value, the backfill, the resolution on every write, and its handling in export, the CLI, draft and cross-workspace deploys, forks and the workspace compare. The run endpoint runs as the caller, so an operator or reader runs an agent with their own access. The editor tells the author of a folder agent that anyone running it needs access to its AI resource and to what its tools use. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * docs: agent editor comments describe runs as the caller Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * refactor: agent editor pane notes use Alert Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * refactor: agent editor pane notes render as Alert Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: agent editor notes as regular Alerts, not banners Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: agent path as its own editor level, and evals on the agent page The path leaves the agent form: the editor dialog opens it as a level, as it does evals, and the editor page in a drawer. The agent page gains Evals, in a dialog. The page supplies the run form, keeping it out of the editor the flow editor reaches. The draft edit gate no longer throws when a focused, changed field is removed: the `change` that removal fires lands mid-teardown, so the gate opens just after instead. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: agent settings as the resource editor's header fields, behind a cog The agent's own settings (path, labels, workspace specific, description) open from a cog as the flow and script editors' do, laid out as the top of the resource editor. The folder note is gone. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: agent settings fields and cog, completing the rename Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * feat: evals open as a dialog over the agent editor page The editor page opens an agent's evals over itself, as the agent page does, with the unsaved edits offered to a run; the editor dialog keeps evals as a level of its own. The two pages share the dialog. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: the unreachable model provider note reads like the unreachable agent one Same warning level and wording as the note above it, with the path inline rather than in parentheses that lost their spaces. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: the unreachable model provider note offers editing the agent too Editing the agent to use a provider the reader can access is often the simpler way out; offered when the reader can write the agent, with Unlink and asking for access. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: provider note lists asking for access as its own option Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: shorter provider note, without the header's buttons repeated Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: evals only for those who can edit the agent Evaluating builds datasets and runs against the agent, which is authoring: the agent page and the editor offer it only with write access to the agent. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: agent page actions ordered as the script and flow pages The menu leads and Edit comes last, as DetailPageHeader lays them out. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: the agent editor dialog's levels slide in built on the first visit Warmed, as the evals pane's levels are, so settings and evals are mounted before the first navigation rather than inside its transition. Evals are only in the strip where they can be opened. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: the agent's settings report path errors, and hold nothing a reader can edit The path field reads its error back, so it shows it and keeps deploy blocked on it. Labels are shown rather than editable without write access. Evals wait for the load to know they can be opened, and the page layout builds no levels it never shows. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * refactor: one builder for an agent's run flow, and the agent page on the shared header The run endpoint and evals build the agent's one-step flow through the same function. The agent page uses DetailPageHeader, whose error handler, tag and trigger context are now optional, and whose menu items keep their disabled state. The home row and the page share the agent's menu and delete. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: the agent page's menu is built from its current path Deploy settings are the workspace's, so they load once and the menu is derived from them rather than fetched per path, where a superseded fetch could land after a navigation. The shared run-flow builder lives with agent runs rather than evals. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: the scoped-read comment names the run-flow builder as it is Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com> |
||
|
ca44043e12 |
feat: let operators compose flows when the workspace grants the right (#11228)
* feat: let a workspace withdraw operator schedule and trigger writes Operators can create, edit and delete schedules and triggers today through the API, CLI and MCP, while the operator_settings flags beside them only hide those pages. An admin who wants operators to see what is scheduled without letting them change it cannot express that. Add manage_schedules and manage_triggers as enforced settings, gated at the schedule handlers and at the generic TriggerCrud routes so every trigger kind is covered by one check. They name capabilities operators already hold, so they are granted unless withdrawn, and absence has to mean "never configured" rather than a value. The read coalesces to true; the update endpoint merges into the stored jsonb with the two fields as Option<bool>, so an omitted key keeps what is stored. operator_settings is git-synced as a whole object, so a settings file written before these keys existed reaches the endpoint on every pull, and a serde or SQL default of either polarity would turn that pull into a silent withdrawal or restoration. The rights are read through a per-process cache, so withdrawing one publishes a notify_event that drops the entry on every replica. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Dsf6VC4MVLisiEoeQkgbr4 * feat: let operators compose flows when the workspace grants the right Adds operator_settings.builder_flows: a workspace setting that lets every operator compose flows out of runnables that already exist. It does not make them authors. The boundary the operator role draws is authoring code and running arbitrary code, and this does not move it: check_flow_is_composition_only walks the value and refuses anything carrying code, including the shapes an obvious walk misses (code hoisted into a flow_node, an AI agent step's tools, and a linked ai_agent resource whose tool list is resolved at run time). What the walk cannot settle it returns for the caller to authorize under RLS: the worker tags the steps pin, every runnable they reference, and the (path, hash) of every version-pinned step. Composing a path is enough to run it and to run it as whoever it runs as, since the worker resolves a step's path with the root DB handle and adopts that runnable's on_behalf_of. A pinned hash needs its own check because dispatch ignores the path beside it. The gate runs on every write and on both request-supplied-value paths, flow preview and flow dependencies, or either becomes the way to run what the write path refuses. Operators of a builder workspace consume a full author seat; the EE companion carries the counting. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Dsf6VC4MVLisiEoeQkgbr4 * feat: enforce operator write rights on the router and in the UI Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: close the capture gap and gate the trigger editors' write actions Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: gate acl writes and the native trigger drawer behind manage rights Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: refuse operator writes with 403 and gate sharing at the drawer Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * perf: resolve identity in the operator write gate only for writes Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: gate the suspended-jobs actions and stop the route check refusing reads Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: explain the empty-state create button when operator writes are withdrawn Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: audit operator settings changes and fold path writes into native rows Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: open locked editors read-only and group the operator settings Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: skip email and azure lookups on editor open while triggers are locked Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * docs: state each operator-rights rationale once in comments Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: address CI review findings on operator write rights Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: keep capture move gated and skip it in the builders while locked Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: refuse builder-rights violations with 403 so operators stay logged in Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * refactor: trim duplication in the operator builder gates Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: hide build app from builder operators on the flow page Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * chore: point at the companion EE PR merged with EE main Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: keep a builder's drafts list loading past drafts they cannot write Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: hide saved agents from builder operators in the step picker Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * chore: note the inlined seat rule and drop orphaned sqlx entries Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: stop a builder's step test from logging them out Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: refuse a builder's dependency job on a path it cannot write Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: keep builders from adding dynamic dropdown code to a flow A flow's dropdown code runs as whoever loads its form, so a builder may keep or drop the code stored on the flow it updates, never add or change it. The builder's editor hides the dropdown types and code, and previews options through the deployed flow; the inline dropdown refusal is a 403 so it no longer logs operators out. Also trims rationale comments repeated across sites. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: show why saving operator settings failed The seat-cap refusal on granting builder rights explains what to do; the toast now carries the server's message instead of a generic failure. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: check builder flow drafts like deploys and treat dropdown code as code A developer who loads a builder's flow draft in the editor runs its dynamic dropdown code as themselves, so a builder's draft now passes the same checks as a deploy. Dropdown code is refused like step code rather than kept or dropped, which also removes the exact-match comparison that refused builders over whitespace. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: bill a builder workspace's operators as developers on cloud The cloud seat count behind the Premium page, the sidebar usage and the fork cap still weighed every operator at half a seat, while the builder right makes them authors. The out-of-repo invoicing job must follow the same rule. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: check a builder's flow draft as it will be stored Draft storage strips NUL escapes after the builder check, so a key ending in one (value\u0000, x-windmill-dyn-select-code\u0000) passed the check as an unknown field and was stored under its plain name. The check now reads the sanitized text. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: list a builder's flow drafts and hide hub imports from builders A builder's undeployed flows now appear in the home list, the flow list and the folder counts. Hub project imports and templates bring scripts and apps along, so builders are no longer offered them. The docs record builders' JavaScript expressions as an accepted risk. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: keep the stored builder right when a settings payload omits it A git-synced settings file written before the key existed withdrew the right on every push. builder_flows now follows the manage_* rights: an omitted key leaves the stored value, and the CLI does not count it as a difference. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: word builder refusals by what the flow contains, test the tag refusal A builder refused on a developer's flow never changed its code, so the refusals now describe the flow ("has inline code, so only a developer can edit this flow") rather than an authoring attempt. The grant confirmation uses the neutral dialog: granting changes billing but destroys nothing. The integration test pins the refusal of a worker tag the workspace cannot use. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: read builder rights from the operating workspace, gate the flow page's audit logs entry Builder rights now come from useOperatorBuilderFlows(), next to the schedule and trigger locks, so an editor embedded for another workspace answers about that workspace; the legacy AI chat, one instance for the whole app, reads the navigation workspace. The flow page's Audit logs entry follows the operator audit_logs setting now that builders open that menu. Operator settings reset every value on load, null settings included, so nothing carries over from the previous workspace. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: pick a dynamic dropdown's code source by the operating user's role The flow input editor, the flow test panel and the flow chat send the dropdown request to the operating workspace, so they now also choose inline versus deployed code by the role held there, through useOperatingUser(), instead of the navigation workspace's. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * chore: update ee-repo-ref to 40ac1c5f8cbce3843b582d9b392d3f3cc7eca3e6 This commit updates the EE repository reference after PR #815 was merged in windmill-ee-private. Previous ee-repo-ref: 31c9e66884b8ca805b20bbfad41fc428fbedbc0e New ee-repo-ref: 40ac1c5f8cbce3843b582d9b392d3f3cc7eca3e6 Automated by sync-ee-ref workflow. --------- Co-authored-by: Claude Opus 5 <noreply@anthropic.com> Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com> |
||
|
|
c1b59f70dd |
feat(ai-agent): add compaction memory that summarizes older context (#10928)
* feat(ai-agent): add autocompacted memory that summarizes older context Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): compact on final-answer turns and count what a turn appended Address the pre-push review findings on the compaction path: - A turn the model answers without a tool call left the agent loop on its first iteration, so a chat-shaped step never compacted and reloaded the whole conversation on every later turn. Compaction now also runs after the loop. - The trigger measured only the last request, so a single large tool result could carry the next one past the window without ever crossing 80%. - The summarization call re-sent the usage-tracking request shape on endpoints the loop had already learned to drop it for. - The flat 8000-token summary reserve swallowed the whole target on a small context window, leaving one message in the tail and summarizing the rest. - A response cut off inside the <analysis> scratchpad was accepted as a summary. - The chat-mode memory default was a shared object the step form edited in place. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): keep Anthropic prompt counts and compact once per response Address the first CI review round on the compaction path: - Anthropic's streaming parser dropped `message_start`, the only event carrying the prompt-side counts, so a native Anthropic run reported no input tokens at all and compaction fell back to a character estimate. - A loop that exits without issuing another request — a structured-output turn does — reached the post-loop pass still holding the previous measurement and compacted a second time, or retried a failure with nothing changed. - The summarization call inherited the step's `max_completion_tokens`; a low one truncates the summary inside its scratchpad, which counts as a failure and disables compaction after three of them. - A fired trigger that found nothing to summarize said nothing. - Memory already over the window — a lowered `context_window`, or a step moved over from `auto` — had no way back, since compaction only ran after an accepted request. It now also runs once before the first one. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): state the summary's own completion cap and drop the pre-flight pass - The summarization call asked for no completion cap at all, which is "uncapped" only on the OpenAI-shaped providers: Anthropic substitutes 64000, over several Claude models' output ceiling, and Bedrock leaves the model's own small default, short enough to cut the response off inside its scratchpad. It now asks for the reserve the split already set aside, raised to the step's cap when that is larger. - Compaction no longer runs before the first request. The fallbacks the loop learns from a rejection are not known that early, so on exactly the endpoints that need them the summarization was malformed by construction: it failed, spent a strike, and the first agent request still carried the oversized conversation. A memory already past the window is repaired on the turn after a request the endpoint accepts, rather than by a pass that cannot succeed there. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): ask the summary for exactly the room the split reserved The split scales its reserve down on a small window while the request asked for a flat 8000, so the two diverged below an 80k window: on a 4k/8k model the cap alone exceeded the window and every summarization was refused, and on a 20k one a full-length summary could land the conversation back over the trigger and compact its own previous summary on the next response. Both now read one `summary_reserve_tokens`. The call also no longer inherits the step's reasoning effort. Every provider counts thinking against that same budget, so a high-effort model could spend the whole reserve before writing anything and return a summary cut off inside its scratchpad; the compaction prompt asks for an `<analysis>` block, which is the reasoning this call needs. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): charge the compaction budget for tools and the system prompt The tail budget was the whole target, but a request also carries the system prompt compaction keeps and the tool definitions, which are not in the message list at all. On a small window those are most of it: a tail sized to the full target left the next request back over the trigger, compacting again every response, and the no-usage estimate missed the tool schemas entirely so it could fail to trigger at all. Both now account for them. The reserve also gains a floor. It is the summary's output cap as well as the room the split leaves, and scaled down without one a small window gave a structured nine-section summary a few hundred tokens — truncated inside its scratchpad every time, which is discarded, which switches the mode off after three. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): count Gemini's tool-use prompt tokens in an agent step's usage Gemini splits a tool-using turn's input across `promptTokenCount` and a disjoint `toolUsePromptTokenCount`, and its thinking apart from `candidatesTokenCount`. The agent step's parser read only the headline fields, so every tool-using turn under-reported both — and the compaction trigger, which runs off the reported prompt, could not see the tool results that grew it. It now goes through the same helpers the proxy path already used. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): calibrate the compaction estimate against the measured prompt Two rounds running, the finding was "the character estimate cannot see input X" — tool schemas, then S3 attachments, which are short paths in the message list and whole images by the time a provider counts them. Enumerating those is a list that only grows, so the estimate is now scaled to the one number that is ground truth: what the provider charged for the last request. Attachments, tokenizer drift and whatever comes next fall out of that, because the estimate is only ever used relative to itself. Also stop the Gemini helpers turning an absent count into `Some(0)`. Downstream, absent means "fall back to estimating the conversation" while zero reads as an empty prompt and would hold the trigger below its threshold for the whole run. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): charge attachments what they cost and let a heavy short prefix compact The calibration conserved the conversation's total cost but spread it by character count, so an attachment — a short S3 path in the message list, a whole image or PDF once a provider expands it — was charged to the text messages around it and stayed nearly free in the split. It now carries a nominal cost of its own, which the calibration corrects a residual on rather than the whole gap. The four-message minimum also refused exactly the case that fix is for: an attachment arriving on the first or second turn can pass the trigger before four removable messages exist, and summarizing even one of them saves most of the prompt. A prefix worth a quarter of the window is now enough on its own. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): never summarize a prefix holding only a previous summary The message-count floor was carrying a second job: a fresh summary sits in a one or two message prefix, so requiring four declined it. The share threshold added last commit admits it, and a summary is reserve-sized by construction — so the post-compaction shape could spend one summarization per response swapping a summary for another the same size, shrinking nothing and losing fidelity each time. A previous summary no longer counts towards that threshold. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): take the context window from the model and drop the estimate calibration Brings compaction in line with how the AI session does the same job, which had already answered these three questions. - The window is looked up from the model. `MODEL_CONTEXT_WINDOWS` in `windmill-ai/src/model_context.rs` mirrors the session's table in `copilot/modelConfig.ts`, entry for entry and with the same matching rules; each side points at the other, since a model added to one and not the other compacts at two different sizes. A step's `context_window` becomes the override for what the lookup cannot serve, and chat mode writes none. - Provider usage is normalized where the provider's quirk is, not at the consumer. `TokenUsage::with_cache_beside_input` raises `input_tokens` to the whole prompt for Anthropic and Bedrock, which report their cached prefix beside it; the OpenAI shape already counts it inside. `prompt_tokens()` is then just `input_tokens`, rather than inferring the shape from whether a write count is present. - The estimator is no longer calibrated against the measured prompt. The session uses the provider's count when it has one and a chars/4 estimate otherwise, with nothing in between, and a tail sized a little wrong only compacts again a turn later. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * feat(ai-agent): summarize memory down to what the database can store Without an instance object store, memory is a 100KB database row cut from its oldest message, the summary included, so compaction on a mainstream model never got to keep anything across runs. A step that persists there now runs its post-loop compaction pass against the smaller of the model's window and the cap at chars/4, about 25k tokens: the loop keeps the whole window, and what is written is a summary plus a tail that fits. The run logs when that pass summarizes, and how many messages the write dropped when one still overshoots. The editor's storage warning on the option is removed: nothing exposes the instance storage to it, so it keyed on the workspace S3 setting, which is unrelated to where memory goes. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): get a complete, billed summary out of every provider Compaction against the real providers turned up four things the stub could not: Gemini and OpenAI's reasoning models think by default and bill it against the same cap the summary must fit in, so the summarization request now asks them for their least (none, low); an OpenAI Responses call that hits max_output_tokens ends in response.incomplete, whose usage the parser dropped, so that summarization went unbilled; a summary that quotes </summary> when it describes its own instruction was cut off at the quote, on the agent step and the AI session alike; and the prefix could end on an unanswered user message, after which the instruction reads as part of that turn (Anthropic merges the two outright). The tail now starts on a user message, and both prompts tell the model the instruction is not part of the conversation. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): compact down to half the window, on the agent step and the AI session The gap between the 80% trigger and the target is what one compaction buys, and every summarization request carries most of the window. At a 70% target a 128k model summarized about 13k tokens of prefix for a summary of up to 8k, so each ~100k-token request bought a few turns of room before the next one re-summarized the previous summary. At 50% the same request frees about 30k. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): drop the workspace-S3 memory hint and state the database bound in the tooltip The memory field warned that memory is kept in the database whenever the workspace had no S3 storage. That setting has no bearing on where memory goes: the instance object store decides, and nothing exposes it to the editor. The field's tooltip now describes both memory kinds and states the database bound unconditionally; the run log says what happened. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): send the summarizer its tool history as text The summarization request carries no tool definitions, and Bedrock's Converse API rejects toolUse/toolResult blocks that arrive without them, so on Bedrock every summarization of a prefix holding a tool call failed silently until the breaker tripped. The prefix's tool calls and results now reach the summarizer rendered as text, on the agent step and in the AI session's compaction, which goes through the same proxy. Also drops the TokenUsage::prompt_tokens accessor, which had become a plain read of the normalized input_tokens, and shortens the context window field's description. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): price attachments from the provider count, bound storage in bytes, effort per pro model Addresses two Codex rounds and a leftovers audit. - Attachments were priced at a flat 1500 tokens in the split, so a multi-page PDF (tens of thousands of tokens to the provider, a short S3 path in the message list) could be kept in the tail or leave no prefix worth summarizing. They are now priced from the provider's count for the request that carried them, less that request's text, with the 1500 floor where nothing was counted. - The database storage bound measured the provider's token count, but the 100KB cap is bytes and repetitive text packs several characters per token. The persist pass now measures the serialized conversation. - The summarizer forced `low` on every reasoning model, which the pro variants reject (gpt-5-pro takes only high, gpt-5.2-pro starts at medium); they now get no effort. - Dropped the unused prompt_tokens accessor and its orphaned assert, an unused PartialEq, a needlessly public lookup, and fully-qualified Gemini calls; refreshed stale comments and the memory_id schema doc; regenerated the flow schema artifacts. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): evict a heavy attachment into the summarized prefix, not the tail Pricing attachments from the provider count was not enough on its own: a leading attachment is a user message, and the boundary rule pulled the last unanswered user turn back into the kept tail to keep it with its answer. For a heavy attachment that dragged it into the tail — or, at the front, emptied the prefix — so it was never summarized and rode every request. The boundary now moves forward instead, keeping that user turn and its answer in the summarized prefix. Verified on the running instance: a 25k-token PDF on a 30k window is summarized out on the turn it overflows, and later turns drop from 26k to ~1.5k tokens. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): keep the forward boundary move off tool results and the prefix start The forward move that keeps an unanswered user turn out of the tail had two edges the third Codex round found: advancing past the user could land the boundary on a tool result (its tool_calls then summarized away, orphaning it), and with no system prompt the summarizable prefix starts at 0, so a trigger firing while the tail estimate fit everything indexed below the start and panicked the task. The forward scan now skips tool-opening boundaries, and the move is guarded above the prefix start. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): drop the step temperature from the summary request OpenAI's reasoning models (gpt-5-mini, gpt-5.1, gpt-5.2) reject `temperature` alongside any reasoning effort but their own default, so a step configured with a temperature made every summarization fail once the summarizer forced a low effort — history then grew unchecked. The internal summary call now omits the step's temperature: a structured extraction does not need a set one, and omitting it sidesteps each provider's temperature-versus-reasoning rules. Confirmed against the API that low + temperature is refused on those models. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): compact an oversized loaded memory before the first request Compaction was reactive, taken only after a request the endpoint accepted, so the fallbacks the loop learns from a rejection are known first. But a memory loaded from an earlier run can already exceed this run's window — the step was switched to a smaller model, or a run under a wider one persisted more than fits — and that first request then overflows and fails the run, with every retry reloading the same history and failing again. A pass is now taken up front, off the character estimate, before the first request. It uses the default request shape; an endpoint needing a fallback may reject this one summary, which is non-fatal, and mainstream providers need none. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): under the storage bound, trigger on the max of bytes and model tokens The storage-bound pass measured only the serialized row size, so an attachment — a few bytes as an S3 path but nearly the whole model context — read as tiny and the pass skipped a compaction the model needed. It now takes the larger of the byte measure and the model's token count, since repetitive text is few tokens but many bytes and an attachment is the reverse; either being over must fire a pass. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): drop oldest turns when a summary cannot fit the window, as the AI session does An oversized loaded memory (a step switched to a smaller model, or an object-store run that persisted more than a later model's window holds) left a prefix larger than the summarizer's own window, so the summary request overflowed and failed, the memory was untouched, and every retry failed the same way. The AI session handles this by falling back from summarization to dropping the oldest turns down to the target; compaction here now does the same. When a summary cannot run — it failed, the breaker is tripped, or nothing is worth folding — the oldest turns are dropped until the conversation fits and opens on a user message, keeping the newest turn. The next request then always fits. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): drop whole turns only, keep the storage pass to bytes, refresh the count after a rewrite Three edges the seventh Codex round found, all in the drop-oldest fallback and the storage-bound measure: - drop_oldest_to_fit dropped to any point that freed enough, which could strand a tool result whose tool_calls went with the messages before it. It now drops whole turns only, always landing the boundary on a user message and never splitting the newest turn; a lone turn too big for the window is left whole rather than broken. - The storage-bound pass measured the whole model prompt against the shrunk 25k window, so a large tool roster and the system prompt — neither written to the row — tripped it on a conversation the row easily held. It measures the serialized bytes alone now; the model's own window is enforced by the in-loop passes and the pre-first-request pass, so the persisted size is all this pass is for. - A compaction rewrites the message list, so the provider's count for the request that produced it no longer lines up. The count is now cleared after any pass that rewrites the conversation, so a later pass measures the estimate over the actual messages instead of a stale, larger prompt (which could decline a summary that already fit and then drop it). The step temperature, no longer sent to the summarizer on any path, is dropped from the request struct. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): measure only the persisted messages against the storage cap Persistence strips the system prompt before writing the memory row, but the storage pass was serializing every message including it, so a large system prompt with a tiny conversation reported far over the storage trigger, and the fallback dropped the one real turn, run after run. The storage measure now serializes only the non-system messages, matching what the row actually holds. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix(ai-agent): run the model-window pass before the storage-bytes pass post-loop A turn the model answered without a tool call broke before the in-loop compaction check, so on database-backed memory its only pass was the storage one, which measures bytes. An attachment fills the model context but is a few bytes in the row, so that turn never compacted and a follow-up could overflow the model. The post-loop now runs a model-window pass first, off the provider's count, then the storage-bytes pass when the row is smaller than the model — both limits enforced for a chat-shaped step, not just the one that happens to bind. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix: simplify agent compaction and preserve execution history * fix: remove unused compaction history setting * fix: preserve answers and recover rejected agent context * refactor: make agent compaction transactional * fix: skip agent summaries that cannot fit retained context * fix: explain skipped agent context compaction * fix: retain recent agent memory when storage compaction cannot fit * fix: start retained agent memory at a user turn * fix: reject unsafe agent memory truncation on storage fallback * docs: clarify agent context window override scope * fix: keep recent turns verbatim when compaction memory outgrows storage Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix: keep the compaction summary out of the agent's answers Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * fix: shorten the agent context window help text Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH * chore: update ee-repo-ref to 942d4013f36edac1fc9a9addbdb02198db1c7a05 This commit updates the EE repository reference after PR #812 was merged in windmill-ee-private. Previous ee-repo-ref: 8ca1682ce6106ba6ea96894fbe606dac64102eb6 New ee-repo-ref: 942d4013f36edac1fc9a9addbdb02198db1c7a05 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> Co-authored-by: Ruben Fiszel <ruben@windmill.dev> |
||
|
hugocasaandClaude Opus 5
|
1d119e6b25 |
fix: correct the tool controls and name field of a nested AI agent (#11221)
* fix: hide tool controls a nested AI agent cannot use Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * refactor: drop the tool navigation prop an agent tool now implies Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: ask for a tool name on every kind of agent tool Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: keep enabled_tools reachable on a linked nested agent tool Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * docs: name the tool kinds the header's name field actually renders Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * docs: state the tool-name rule without listing the kinds it covers Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 <noreply@anthropic.com> |
||
|
AlexRV12andClaude Opus 5.5
|
1b74939952 |
feat: support aiagent steps in test_run_step (#11321)
* fix: support aiagent steps in test_run_step Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: offer the agent's tools as a picker in the step run form Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * fix: run a blank tool list as no tools when the step form is bypassed Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com> |
||
|
hugocasa
|
d8aaa73573 |
feat: chat with a saved agent from the agent editor (#11292)
* feat: chat with a saved agent from the agent editor * fix: keep agent chats apart from a same-path flow's conversations * fix: point an agent editor chat's model gap at the agent, not a step * fix: match the memory gate's icon to the chat's empty state |
||
|
6186a0645d |
feat: edit variables, resources and triggers in their own session tab (#11206)
* feat: edit variables, resources and triggers in their own session tab Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: take a trigger's flow kind from its config and name the written trigger Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: one operating-workspace context for editors acting on a session's workspace Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: reload page item tabs only in sessions acting on the tool's workspace Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: resolve open in workspace for every preview tab kind in one place Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: guard every component under a session editor against the navigation workspace Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: show a page item tab whose item is gone instead of mounting its editor Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: follow a saved page item in the tab's own workspace and restore in-frame rows Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: close a page item tab from its editor header Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: judge trigger permissions by the user acting in the session workspace Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: judge every session-editor permission by the user acting in that workspace Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: keep session permissions reactive to the acting role and restore tabs as last seen Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: default Path's owner controls to the user acting in its workspace 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: Ruben Fiszel <ruben@windmill.dev> |
||
|
hugocasaandClaude Opus 5
|
a08992834d |
feat: render an AI agent result as its answer, not as raw JSON (#11051)
* feat: render an AI agent result as its answer, not as raw JSON Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: sanitize agent markdown through the shared plugin chain Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: fold agent stream events incrementally per poll Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: separate the agent meta line from the result toggle group Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: add a transcript view of an agent run's conversation Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: retire AIAgentLogViewer in favour of the transcript Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: show the partial transcript a max-iterations failure carries Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: move the agent meta line and system prompt below the conversation Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: show an agent run as what it did, not as a conversation Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: name the agent run breakdown a trace Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(ai-agent): label the save-as-agent form fields per the guidelines Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(ai-agent): reuse the resource form's path and description fields Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(ai-agent): keep the action tags on a max-iterations failure Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(ai-agent): keep offline replay inert and the streamed answer to one turn Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(ai-agent): reset the streamed answer on providers that skip tool_call Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: carry the event type narrowing through the stream parser Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: survive a malformed message rather than take the viewer down Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: coerce agent messages once at the parse boundary Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: show an agent run as one scroll ending in its output Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: keep every streamed turn instead of dropping the narration Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: share the chat divider and drop the unsafe run auto-scroll Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: give the labelled divider a border colour and the standard pretty icon Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: pad the agent run below its badges as well as above Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: follow a streaming run's pane without moving the page Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: share the chat's stick-to-bottom mechanics with the agent run Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: keep the answer's citations and end a turn's reasoning with the turn Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: read the agent stream through the windmill-chat sdk parser Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: show an agent run's thinking instead of falling back to raw json Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: stop a stream the fold cannot use from claiming the run pane Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: identify an agent run by more than the job id the replay withholds Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: drop the tests and comment lines that were not earning their place Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: stop a streamed turn's text shifting when a tool call closes it Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Guilhem
|
e2a91ca2b1 |
feat: flow chat job-backed detail, smooth streaming and answer chrome (#11186)
* feat: flow chat job-backed detail, smooth streaming and answer chrome * chore: document setFlowPath, drop unused chatIdentity, tighten comments * fix: read a nested agent tool's call from its own job * fix: show only the model's arguments on a tool card read from its job * feat: add attachments to the chat sdk message type * refactor: read a flow chat message's files from the message, not its run * fix: keep every job-backed answer when reads finish out of order * refactor: build flow chat tool cards from the conversation row alone * feat: keep the model picker's current choice when a flow chat retry replays its run * fix: export ChatAttachment from the chat sdk entry point * fix: drop a conversation list for the flow path before setFlowPath * fix: return the current listing, not old-path rows, after setFlowPath * fix: treat a listing stale on kind or path as one stale listing * revert: drop setFlowPath and rebuild the chat when the flow path changes * fix: refuse a flow chat retry once any turn ran while it read the run * docs: say where a display message's createdAt comes from * style: drop the rule between New chat and the conversation list |
||
|
GuilhemandClaude Opus 5
|
c4c9677982 |
feat: attach files to a flow chat message (#11185)
* feat: attach files to a flow chat message Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: store chat uploads under windmill_uploads and withdraw refused sends cleanly Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: refuse extra files for a single-file input and keep attachments on retry Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: clean up partial upload batches, withdraw a stopped send once, free retry payloads Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: list a chat conversation only once its run starts and require files where the flow does Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: carry uploaded attachments on the pending chat user message Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: route rich-composer file paste through the attachment lanes and tighten comments Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: keep a stopped turn's files for retry and let an explicit media type win Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: refuse chat attachments sent without text and shorten comments Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: resolve the attachments input from real flow input reads, ignoring loop iteration Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: read flow input references through one parser for the model and attachments controls Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: withdraw an attachment send stopped after its uploads answered Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: discard uploads when a send is stopped as its attachments are announced Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs: keep uploadAttachments' doc comment on uploadAttachments Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: never delete chat uploads from workspace storage Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * test: pin that a failed upload aborts the rest of its batch Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
hugocasaandClaude Opus 5
|
c297ed0052 |
feat: managed memory with an inherited or custom memory id per step (#11118)
* feat: split ai agent memory into agent policy, run memory id and step history * fix: scope string memory ids to workspace and flow, keep nested tool history inputs * chore: update sqlx cache for the flow context query * docs: describe memory id scoping as collision-free rather than isolated * chore: regenerate openflow json after merging main * fix: offer no memory id for legacy manual memory, document linked history inputs * fix: seed provided messages from legacy manual memory and hide its note once set * fix: bypass memory when a provided messages expression evaluates to null * fix: require a user message when provided messages are empty * chore: keep the empty messages comment within the line width * docs: name the history inputs wherever linked steps list their flow-local inputs * docs: keep the memory storage path on one line * feat: managed memory with an inherited or custom memory id per step Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: list a custom memory id in the test run form and name where an inherited one comes from Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: keep memory id out of the add-field menu and drop the memory id telemetry Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: keep legacy auto memory without an id working after an untouched redeploy Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: rename step messages to previous_messages and address review Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * style: rewrap comments and docs lines lengthened by the previous_messages rename Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * refactor: read agent memory as either a legacy shape or the current one Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: name the memory setting in ignored-input notes and keep conversions honest Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: keep a legacy memory count unset on open and read a cleared count as off Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix: address review on cleared test history and zero-count memory Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix: drop flow-local keys from a linked agent resource before interpolating it Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix: keep a linked resource's own inputs as fallbacks and note ignored history on image runs Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix: restore the linked agent draft tests and log ignored history on every image run Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix: resolve the one-of variant from the value when the selected one leaves the list Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix: treat zero-count managed memory as off when enabling chat mode and shorten comments Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix: stop requiring user_message in the openflow agent contract when previous messages are the prompt Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 <noreply@anthropic.com> |
||
|
GuilhemandClaude Fable 5.1
|
4eab995cf7 |
feat: tell test flow conversations from deployed ones and rename a chat (#11179)
* feat: mark test flow conversations apart from deployed ones and allow renaming a chat Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix: keep the conversation kind across refreshes and reject NUL titles Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix: ignore conversation lists for a kind no longer selected Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: start a fresh conversation listing when the kind changes on a later page Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: refuse sending into a conversation of the other kind Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: refuse cross-kind conversation continuations on the server Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: keep the conversation filter unavailable while an answer runs Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
GuilhemandClaude Fable 5.1
|
189793c2e4 |
feat: flow chat model picker on a shared model-settings component (#11187)
* refactor: render the session chat model menu from a shared ChatModelSettings config Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * feat: pick the flow chat's model and thinking from the provider fields the flow exposes Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix: name only the thinking level the flow run will send on the model button Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: let the flow chat take a typed model id and keep a shared thinking input editable Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: promote a flow input to the model button only where its control can edit it Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: drop any reasoning token the chosen model rejects before a flow chat run Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
GuilhemandClaude Fable 5.1
|
68f2248018 |
feat: keep flow inputs and seed the agent when chat mode is enabled (#11177)
* feat: keep flow inputs and seed the agent when chat mode is enabled Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix: leave a linked agent's memory and streaming to the agent when enabling chat mode Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> --------- Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
hugocasa
|
57a99f66a8 |
feat: rename saved agents from the agent editor and flag broken links (#11147)
* feat: list flows that link a saved agent and flag broken agent links * feat: rename saved agents from the agent editor and repoint the flow * fix: show an unreadable linked agent as not accessible, not missing * fix: address review nits on agent rename and missing-agent state * fix: open content search above modals and keep Escape for it * fix: register content search on the opener's overlay stack * docs: scope the global search z-index comment to the bases it clears * refactor: show linked agents' rename warning as for scripts and flows * fix: keep the failed-lookup rename warning to resources |
||
|
Ruben FiszelandClaude Fable 5.1
|
8b4f6220dc |
refactor: run the flow chat UI on the windmill-chat sdk (#11134)
* refactor: run the flow chat UI on the windmill-chat sdk Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix: structural answer check, poll option and latest run in the chat sdk Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix: count jobless tool rows and the loaded license in the flow chat Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> --------- Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
|
a78beff743 |
feat: dynamic AI agent toolsets (#11050)
* feat: dynamic ai agent toolsets, and memory as a step input Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: address review round 1 on dynamic ai agent toolsets Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: tag enabled_tools and drop the memory step input Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: let an mcp server entry be named by the path the roster shows Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: keep $res: out of the tool names the enabled tools picker offers Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: name an mcp server by its bare path on the one side that can hold it Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: count the enabled tool names that matched nothing instead of logging them Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: narrow an agent's roster in one pass, by whole entries Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * test: pin that an mcp summary is rejected against a name that is not Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * chore: regenerate the copilot flow schema after the merge Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs: shorten the enabled tools list hint Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: take enabled_tools back to a plain list of tool names Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs: keep the enabled tools add-menu hint describing the unset field Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: name a websearch tool that carries no summary of its own Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: reserve the name web search is enabled by so no tool can share it Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: spell the reserved web search name with a hyphen Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: reserve __wm_web_search as the name web search is enabled by Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * chore: advance ee-repo-ref past the git sync ci check work Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs: shorten the enabled tools description the run form shows Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * chore: update ee-repo-ref to e4c1b794d6c5e6e390987341b2840587bbb40348 This commit updates the EE repository reference after PR #785 was merged in windmill-ee-private. Previous ee-repo-ref: af668462f0f06b02a5f4e0c22e6156858487a518 New ee-repo-ref: e4c1b794d6c5e6e390987341b2840587bbb40348 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> |
||
|
|
63cb46d7bb |
feat: add a minimal skin for the approval page and slack/teams (#11061)
* feat: add an approval skin to the approval page and slack/teams messages Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01KK2Ye4PizykReZVuMEm3m9 * chore: point ee-repo-ref at the teams approval skin commit Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01KK2Ye4PizykReZVuMEm3m9 * fix: resolve the approval skin from the step awaiting approval Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01KK2Ye4PizykReZVuMEm3m9 * fix: shorten the slack approval message to fit the button value limit Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01KK2Ye4PizykReZVuMEm3m9 * fix: rename skins to detailed/minimal and keep long slack messages Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01KK2Ye4PizykReZVuMEm3m9 * feat: title the minimal approval page from the step and flow summaries Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01KK2Ye4PizykReZVuMEm3m9 * feat: let wait_for_approval set the description approvers see Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01KK2Ye4PizykReZVuMEm3m9 * fix: keep a finished workflow's approval description, still gated Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01KK2Ye4PizykReZVuMEm3m9 * fix: keep a login-required approval locked after the run moves on Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01KK2Ye4PizykReZVuMEm3m9 * feat: hide the windmill version on the approval page Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01KK2Ye4PizykReZVuMEm3m9 * chore: update ee-repo-ref to e92abc9d1fba3ba898640df0cfeb0af8a50849b4 This commit updates the EE repository reference after PR #786 was merged in windmill-ee-private. Previous ee-repo-ref: bf1766ff49458f62d3f11746f1a06893ae3c2325 New ee-repo-ref: e92abc9d1fba3ba898640df0cfeb0af8a50849b4 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> |
||
|
hugocasaandClaude Opus 5
|
7feaf619cf |
feat: run a linked AI agent's draft when testing a flow, and offer to deploy it (#10993)
* feat(frontend): run a linked agent's draft when testing a flow, and offer to deploy it Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * fix(frontend): settle an agent's autosave before reading it, and refresh its card on a draft save Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * fix(frontend): deploy the agent draft that was validated, and make the draft-tools flag explicit Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * fix(frontend): refuse a stale agent deploy, and warn when a never-deployed agent is kept as a draft Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * docs: record what inlining an agent draft puts in a preview job Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * fix(frontend): name the draft-changes dialog after what it lists Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * fix(frontend): refuse a draft deploy when the draft row is gone Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * fix(frontend): apply the missing-draft refusal to raw apps too Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * fix(frontend): stop reading a deployed resource row as a draft on deploy Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * fix(frontend): do not mistake an outage or a vanished draft for a deploy Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * fix(frontend): refuse an agent read whose pending draft save failed Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * fix(frontend): space the trigger badges and right-align the agent actions Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * fix(frontend): surface a failed agent-draft read instead of dropping it Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * fix(frontend): give the agent draft delete a baseline so a newer edit survives Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * fix(frontend): drop the agent draft cell locally instead of deleting twice Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * test(frontend): pass the withDraft flag the guard tests were missing Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * refactor(frontend): deploy agent drafts the way Review & Deploy does Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * fix(frontend): give the read-only flow graph its own linked-tools bucket Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * fix(frontend): base the resource draft delete on the read that promoted it Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * fix(frontend): refresh every step linking an agent when its draft is saved Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * fix(frontend): write nothing at all when a resource draft has gone Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp * test(frontend): pin that the resource draft delete follows its baseline seed Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017B4omp8dRgmLitbpQEqFMp --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
hugocasaandClaude Opus 5
|
f037c73d10 |
feat(frontend): group the agent form and edit saved agents as drafts (#10880)
* feat(frontend): group the AI agent step form and edit saved agents in a modal Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat: edit a saved AI agent through its own resource draft Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * refactor: drop the agent fork-for-edit session now that edits live in a draft Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat: edit ai_agent resources from the resources page with the agent editor Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: send a standalone agent's brain from the module when testing a step Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: keep the agent draft faithful to the resource it deploys to Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * chore: add the sqlx cache entry for the eval subject rename Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * refactor: share the module insert between the graph and the agent editor Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat(frontend): open evals inside the agent editor, actions in its header Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat(frontend): add tools from the agent editor and lighten its test pane Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(frontend): open an ai_agent deep link in the agent editor Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(frontend): drop the failed result badge on a step that never ran Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat(frontend): head the agent editor's levels with a back control Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat(frontend): drop connect and fill inputs from the agent editor Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat(frontend): lighten the agent editor's run panel Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(frontend): stop a nested agent tool's config reading as AI-filled Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat(frontend): offer only AI or static on an agent tool's inputs Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat(frontend): give a saved agent's tool editor a static-only surface Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat(frontend): open an agent tool in a drawer beside the agent Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat(frontend): hide unset agent config in the run form Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * refactor(frontend): share the input forms' pickers and s3 lookup Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * chore(frontend): drop a dead agent-editor export and fix two stale comments Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(frontend): reach an ai_agent's resource-level settings and copilot Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(frontend): open an ai_agent's resource view as JSON, not the generic form Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(frontend): address review findings on the agent editor's draft and streaming Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(frontend): close the agent editor on a version restore, as the resource editor does Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(frontend): stop the provider picker auto-writing a kind, and clear review nits Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * refactor(frontend): drop the fork-for-edit leftovers from the agent card Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(frontend): mount the agent editor in the dev flow editor and guard the deep-link race Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(frontend): deploy the agent config that was submitted, and refuse one no run could use Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * refactor(frontend): build the agent editor's rows from the design-system button Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(frontend): keep a draft-only agent's draft, and let a blank MCP summary deploy Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(frontend): guard read-only agents, incomplete MCP tools and duplicate editor mounts Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: read-only agent editor, linked-card refresh, atomic eval rename * fix: eval rename needs the privileged pool, per-workspace write access * fix(frontend): drop the agent editor target when its mount goes away * refactor: drop the agent rename work from this PR, unban the bindable defaults * fix(frontend): refuse a renaming deploy and drop the copilot from static-only fields * fix(frontend): mirror the worker's streaming rule and scope agent writes to their target * fix(frontend): read runtime streaming as off and reset the drawer's json view * fix(frontend): read an unsettled output_type as non-streaming too * fix(frontend): let the showing modal claim an agent opened from inside it * fix(frontend): keep in-flight edits, tool replacements and every linked step in sync * fix(frontend): keep attachments in the run form and bind the agent ref to its tools * fix(frontend): preview the agent as authored and re-evaluate step args on run * fix(frontend): scope agent-editor ownership to the flow's workspace * fix(frontend): drop the tool drill-in where there is no graph to select on * fix(frontend): require a provider kind and keep one resource editor open at a time * fix(frontend): keep legacy nulls, static-only text literal, and the handover anchor * test(worker): pin the agent streaming default * fix(frontend): let an AI-fillable input be switched to static * fix(frontend): report agent editor background failures instead of floating them * fix(frontend): keep the version pane's path alive while the editor closes * fix(frontend): clear the anchor-keep flag at the start of each drawer session * fix(frontend): preview the agent without its synthetic path, refresh the baseline on external writes * refactor(frontend): drop the unverifiable baseline refresh, state the synthetic-path rule * fix(frontend): keep the synthetic path out of agent tool test runs too * refactor(frontend): host the agent editor under the agent's own path * fix(frontend): mark an agent editor's host explicitly instead of inferring it from the path * fix(frontend): discard linked-agent responses from before a deploy * fix(frontend): keep a flow mount from claiming an agent editor's nested target * feat(frontend): keep an agent used as a tool inside the agent being edited * fix(frontend): reserve the agent editor's root module id * docs(frontend): record why the agent editor previews under the agent's path * fix(frontend): refuse to open or deploy a resource that is not an agent * docs(frontend): put the scope-migration comment on the function it describes * fix(frontend): refuse an agent path whose resource type is not proven * fix(frontend): recheck the resource type before deploying, and keep expressions off static-only inputs * fix(frontend): lazy-load the agent editor and slide its levels like the evals pane * refactor: drop unreachable non-list tools check from agent deploy Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(frontend): clear text-only agent fields on image output, reserve the root id Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(frontend): keep the agent editor usable for a non-list tools value Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(frontend): stop the parked eval run list from taking arrow keys Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(frontend): report a non-list tools value on deploy instead of throwing Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(frontend): keep temperature editable for image output Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(frontend): skip non-object tool entries when rendering an agent Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(frontend): guard tool entry reads instead of copying the tool array Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(frontend): key tool rows by position so duplicate ids render Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 <noreply@anthropic.com> |
||
|
Ruben FiszelandClaude Fable 5.1
|
fdd3b36423 |
feat: workspace setting to hide the AI assistant, agent steps unaffected (#10941)
* feat: workspace setting to hide the AI assistant, agent steps unaffected Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011eNweUugVqerex6MLxjbeL * fix: load workspace AI config on cold /sessions load and say hidden, not disabled Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011eNweUugVqerex6MLxjbeL * fix: follow workspace switches on /sessions gate and drop deprecated button size Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011eNweUugVqerex6MLxjbeL * fix: key the /sessions hidden-assistant gate on the acting workspace's own config Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011eNweUugVqerex6MLxjbeL * fix: tag the /sessions hidden-assistant verdict with its workspace and drop superseded reads Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011eNweUugVqerex6MLxjbeL * fix: overlay the /sessions hidden-assistant gate so warm sessions survive workspace switches Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011eNweUugVqerex6MLxjbeL * fix: hide the pipeline insert menu AI prompt and refuse chat turns where the assistant is hidden Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011eNweUugVqerex6MLxjbeL * fix: shrink the home Build with AI / CLI / Hub line to a flush hint row Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011eNweUugVqerex6MLxjbeL * fix: frame the workspace toggle as hide AI sessions at the bottom of the AI settings Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011eNweUugVqerex6MLxjbeL --------- Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Ruben FiszelandClaude Fable 5.1
|
95b6bbd46a |
fix: preselect first row of AI agent and AI sandbox insert panes (#10937)
* fix: preselect first row of AI agent and AI sandbox insert panes Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019vjSnhnewkUbx6mR9iCeK8 * fix: keep Enter for focused controls in the AI insert panes Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019vjSnhnewkUbx6mR9iCeK8 |
||
|
|
0bbd559ac8 |
feat: instrument AI fill/fix, evals, agents and the debugger (#10853)
* feat: track AI fill, AI fix, evals, reusable agents and debugger usage Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * chore: pin ee ref to the feature_usage registry commit Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * chore: update ee-repo-ref to c3b6f62ea579a3583d4b474e9885c77104cfc87e This commit updates the EE repository reference after PR #745 was merged in windmill-ee-private. Previous ee-repo-ref: 77992910929188a854eadc06ee45971877b6f954 New ee-repo-ref: c3b6f62ea579a3583d4b474e9885c77104cfc87e Automated by sync-ee-ref workflow. --------- Co-authored-by: Claude Opus 5 <noreply@anthropic.com> Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com> Co-authored-by: Ruben Fiszel <ruben@windmill.dev> |
||
|
AlexRV12andClaude Opus 5
|
6b73145e72 |
fix(frontend): follow the operating workspace in step input forms (#10834)
* fix(frontend): follow the operating workspace in step input forms Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(frontend): rethrow auth errors and wire remaining variable pickers Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor(frontend): use runed watch for picker workspace reloads Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
AlexRV12andClaude Opus 5
|
93081e255f |
fix(frontend): mint string password secrets in the operating workspace (#10815)
* fix(frontend): mint password secrets in the operating workspace A `password: true` string argument is rendered by PasswordArgInput, which mints an ephemeral secret variable on the first keystroke and rebinds the argument to `$var:<path>`. It minted into `$workspaceStore` — the globally active navigation workspace. Session editors operate on a different, possibly forked workspace without switching `$workspaceStore`, and thread that operating workspace explicitly as a `workspace` prop. When the two diverged the secret landed where the user was merely looking while the job ran elsewhere, and the backend failed with `Variable not found`. Add the `workspace` prop to PasswordArgInput and thread it through every hop between a form mount and the minting field, plus the entry points that supply it. Track `mintedIn` so updates target where the variable actually lives, and re-mint when the operating workspace moves after a path already exists. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(frontend): keep a password field consistent with its argument A parent can replace the whole args object without remounting this field — previewing a saved input, say — leaving `path` and `password` describing a secret the argument no longer points at. Minting from them then copies the old plaintext over the replacement, and the replacement is lost. State that rule once as `argReplaced` and gate every mint on it. The replacement can also land while the create is in flight, so the bound value is captured before the request and re-checked after it resolves; the variable that mint produced was never referenced, so it is deleted outright. A mint that ends without binding re-seeds `password` from what the argument now holds, so the field stops displaying a secret that will not be submitted and a later workspace move cannot re-mint the stale plaintext. `updateValue` returns early before anything is minted, since its 404 retry would otherwise bind over a replacement it cannot see. A failed initial mint now raises a toast rather than passing silently, which also removes the component's last unhandled rejection. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * test(frontend): guard workspace forwarding to PasswordArgInput Every hop between the form a caller mounts and the PasswordArgInput that mints the secret must forward `workspace`, and so must the entry points that supply it. A hop that drops the prop falls back to the navigation workspace while the top-level case keeps passing, and no typechecker catches it because every hop declares `workspace?: string | undefined`. The forwarded expression is checked rather than the prop's presence, so `workspace={$workspaceStore}` and `workspace={undefined}` fail. Two ways the scan could stop guarding without failing are asserted too: an unterminated mount raises instead of swallowing the rest of the file, and the number of mounts parsed must equal the number of tag occurrences, so a mount written inline rather than at the start of a line fails loudly. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(frontend): list and create variables in the operating workspace * fix(frontend): surface and bound a failed recovery mint * test(frontend): end a mount at the first line closing it --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
hugocasaandClaude Fable 5
|
9c557859c5 |
feat: AI agent evals: datasets, scored runs and comparison (#10633)
* feat: eval datasets and standalone runs for reusable AI agents Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: agent eval drawer with case editor, runs and capture entry points Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs: document AI agent eval datasets and standalone runs Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: say how many eval cases the list is not showing Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: address review findings on eval datasets - keep an edited case's conversation and tool inputs: serde(flatten) silently drops Box<RawValue> fields, so the update payload is spelled out - remount the case editor per case so one case's turns cannot leak into another - require jobs:read / flow_conversations:read on the capture endpoints, which UserDB does not gate by token scope - take the dataset lock in create and update so a delete cannot be undone by a concurrent metadata write, and delete cases before metadata - load more cases beyond the first page, and stop capping the agent picker - record that the version stamp is taken at enqueue, not at resolution Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: address round-2 review findings on eval datasets - block operators from dataset and case writes - pass the editor's operating workspace through the drawer and the capture request, instead of assuming the navigation workspace - discard superseded case-list responses so switching datasets cannot land the previous dataset's cases - reject a dataset without a case_id (or vice versa) rather than running an inline case under a dangling association - run unsaved edits inline instead of silently running the stored case - surface the API error body on a failed run - fetch dataset metadata concurrently when listing - $bindable() without a default on the optional open prop - correct the permission and enqueue-time-version wording in the docs Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: run an untouched saved case by reference again The editor writes back keys the stored case omits, so comparing the raw objects reported every unedited case as edited: the run went inline and lost the dataset/case stamp its history depends on. Compare a normalized form, and pin it with a test. Also scope the history query to the drawer's workspace and drop superseded responses. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: show a dataset's cases as a table, and fix round-4 review findings The case list showed one case at a time with no overview. It is now a table with the case, where it was captured from, and its last run — the last-run column is a single jobs query on the path stamp rather than a request per row. Review fixes in the same file: - keep the edit baseline on the selected case rather than looking it up in the loaded page, so a case beyond page 1 is not treated as unedited and run stale - release the loading state when a superseded case load returns early - reload every loaded page after a write instead of collapsing to page 1 - last remaining 'resolved to' wording in the version tooltip Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: run a dataset as an experiment, with scorers as runnables An experiment runs every case of a dataset against one subject and records the exact case set it executed, so a result set stays reproducible while the dataset keeps changing. Each case runs as its own small flow — the agent, then a step per scorer — so a case keeps the run stamp, history query and trajectory view a single run already has, and scorers need no orchestration of their own. Results are read back per step by node id rather than by walking a nested loop's status. A scorer is any runnable taking (input, output, expected): a script, a flow, or a reusable agent used as a judge. A judge is prompted with the case and the answer as one JSON message; a script or flow receives them as named arguments. Scores accept a bare number, a boolean or {score}. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: results table for an experiment, with scorer columns One row per case: status, the agent's answer, and a column per scorer, with the mean per scorer above the table and a link into each case's run for its trajectory. Averages skip cases a scorer produced no number for — counting a missing score as zero would read as a regression. The drawer's left pane becomes Cases / Results, and Results carries the scorer picker and Run dataset. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * feat: compare an experiment against a baseline Per-scorer deltas on each row and on the mean, and a filter down to the rows that regressed. Rows join by case id, so a case added after the baseline ran has no delta instead of counting as a change. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: address round-5 review findings on experiments - match scorers by label when diffing two experiments; joining by array position subtracted one scorer from another whenever the scorer sets differed - report a row's status from the case job, not the agent step, so a case whose scorer failed no longer reads as a success - delete a dataset's experiments with it: they hold copies of its cases, and a recreated dataset of the same path would have exposed them - select the experiment that Run dataset just started instead of leaving the table on the previous one - expected is scored now, so stop describing it as having no consumer Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: address round-6 review findings on experiments - hold the dataset lock across an experiment launch, so a delete landing between reading the cases and writing the experiment cannot recreate the deleted dataset's inputs - match scorers between experiments on kind and path, not on label: labels default to a path's last segment, so f/a/quality and f/b/quality compared against each other - average mean deltas over the cases both runs scored; comparing each run's own average reported a regression from a case the baseline never ran, with no regressed row to point at - openapi: the row status is the job's, which is also canceled/skipped; runEval takes scorers; the update-case body no longer advertises source, which the handler deliberately ignores - record why the experiment prefix cannot reach a sibling dataset Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: address round-7 review findings on experiments - release the dataset lock for the push loop and retake it for the write, re-checking the dataset still exists: holding it across the whole launch made every capture and case edit on that dataset 409 until the last job queued - assemble experiment results with bounded concurrency; a 100-case, 3-scorer experiment was 400 sequential lookups, each itself several queries - clear the baseline when it becomes the selected experiment, which was comparing a run against itself and reporting zero deltas - take the header mean over the same cases as its delta while comparing, so the two numbers beside each other describe the same set - a canceled or skipped case is no longer the same grey dot as a running one Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: address round-8 review findings on experiments - verify the dataset's identity, not just its existence, before recording an experiment: the path can be deleted and recreated during the push loop, and the experiment holds copies of the old dataset's cases - give the recording lock a longer budget than a case edit, since its jobs are already queued and giving up strands them, and say so when it fails - keep score lookups sequential within a case: nesting two bounded streams multiplied into 32 in-flight queries against a 50-connection pool - clear a baseline that no longer belongs to the loaded experiments, so switching datasets does not leave comparison mode on with nothing to compare - keep a scorer's own mean when the baseline never ran it, instead of blanking a column full of numbers - EvalCaseDraft.expected no longer claims nothing scores it Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: do not trust an experiment's job ids, and require write to record one Experiment objects live in workspace object storage, which a script can write directly, and results are read on the unrestricted pool — so a forged experiment naming another flow job returned output the jobs API would have refused. Only jobs this server stamped with that experiment's id are read now. Also from round 9: - recording an experiment requires write on the dataset, not read: it persists into the dataset's namespace and its shared list - clear the results table when the selection changes and surface a failed load, instead of labelling the previous experiment's numbers as the new one's - a storage fault is no longer reported as a deleted dataset - the lock-timeout message at the recording site no longer says to retry, which would run the whole dataset again on top of the jobs already queued - ExperimentRow.status documents canceled and skipped Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: bind the experiment trust check to the requested dataset The previous check matched jobs on the experiment id alone, which the stored object supplies — so copying another dataset's experiment JSON under a readable key carried its jobs' output along with it. A job is now only read if it was stamped for this experiment *and* for the dataset the caller's read access was checked against, and an experiment that names a different dataset is not served from this key at all. Also from round 10: - add the .sqlx entry for that query; without it every SQLX_OFFLINE build failed - serve results over GET: as POST the route-scope middleware classified a read as ai_evals:write, locking read-only tokens out of their own results - clear the selected and baseline experiments synchronously when the dataset changes, so the previous dataset's id is not requested under the new one Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: address round-11 review findings on experiments and scorers - give scorers the whole case input, not just the message: an answer that came from attachments or a replayed conversation could not be judged on it - accept a judge's boolean and structured {score} answers, including stringified ones, and pin every documented scorer shape with a test - record an experiment for the cases that did launch when a later push fails, instead of leaving those jobs running with nothing to attribute them to - do not capture a preview parent's synthetic runnable_path as a host flow; the saved case could not be rerun - clear the case table before loading a dataset and surface a failed load, so a failure cannot leave the previous dataset's cases under the new name - keep the results table through a refresh of the same experiment - exclude flow-step jobs from the per-case last-run lookup - drop case sets from the experiment list, which is only used to pick a run - report a database failure at the recording lock as itself, not as contention Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: address round-12 review findings on capture and run history - load flow_node.flow for flownode parents: an agent inside a deployed branch or loop captured without its agent, host flow or tool bindings - decide host_flow_path by whether the path resolves to a flow, not by job kind: excluding previews wholesale also dropped the flow editor's step test, whose path is real - page the per-case last-run lookup by created_before until the loaded cases are covered; one page of 200 reported older cases as never run - do not record an experiment when nothing launched - only attach the case input to a job when a scorer will read it - keep the case table through a save; only a different dataset clears it - drop the superseded duplicate comment on the score parser Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix: stop refetching run history on every case write Reading the case list before the first await made the whole job-history query a dependency of it, so every save, delete and Load more refetched up to 1000 job rows and blanked the column. Read untracked instead. - an empty Last run cell now distinguishes never-ran from not-found-within the page bound, which the comment already claimed and the cell did not - reloading a dataset no longer replaces a populated table with a skeleton - keep the score-parser comment that describes every shape it handles Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * refactor: keep eval datasets in Postgres instead of object storage Datasets, cases and experiments become rows (`eval_dataset`, `eval_case`, `eval_experiment`, `eval_experiment_case`) rather than objects under a `wmill_eval_datasets/` prefix. What a run produced is still the job's: only case inputs and an experiment's case snapshot are stored. This removes the machinery the object store needed: - The advisory lock and the read-modify-write of a per-dataset JSONL. A case is a row, so there is nothing to serialize. - The launch-time identity check on the dataset. The foreign key makes a concurrent delete fail the transaction instead. - The trust guard on an experiment's job ids, which existed because a script can write workspace object storage directly and could forge an experiment naming somebody else's job. An experiment now chooses every job id and records itself before pushing anything, so a launch that dies partway leaves a recorded case whose job is missing rather than a running job nothing accounts for; cases that never reached the queue are removed again. Row-level security on `eval_dataset` is the authority on who may read or write a dataset, so `extra_perms` grants work and the rule is not mirrored in Rust. Cases and experiments carry a read policy derived from their dataset and no write policy: they are written on the unrestricted pool after the dataset row itself has been asked, with `SELECT ... FOR UPDATE`, whether the caller may write it. Cases are capped at 256 KiB each and 10 000 per dataset, refused rather than truncated. Attachments are S3 references, not inline bytes, so a case that approaches either cap is a mistake rather than a use case. Evals no longer need the `parquet` feature or a configured workspace object storage. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * style: align the eval drawer with the design system - Scorer chips are `Badge`s rather than a hand-rolled bordered span, and the section header is a `Label` with its tooltip, as are the case editor's fields (which also gets the label colour right). - The results table showed status as a coloured bullet, which says nothing to a colour-blind reader. It now carries the same icons the runs table uses, with the status as its accessible name. - Feedback colours move to the `-500` shades the brand guidelines name. - The conversation JSON error uses `TextInput`'s `error` prop for the border and the caption style for the message, as elsewhere. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat: author an expected answer, tags and attachments on a case Every scorer is handed `(input, output, expected)`, but nothing could produce an `expected` except a conversation capture: the case editor had no field for it and a captured run left it empty. So: - The editor gains Expected, Tags and a read-only list of the attachments a captured case carries. Expected is plain text, or JSON when the answer has structure. - Capturing from an AI agent run keeps what that run answered, which is the only moment a reference answer exists for free. The results table also laid itself out by content, so a long answer pushed the scores — the numbers the table exists for — off the edge of the pane. It is fixed-layout now, with the text columns bounded. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * docs: expected is captured from a run and can be authored Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat: link a saved agent when inserting an ai agent step "AI Agent" in the step picker was a leaf that always created a blank step, so reusing a saved agent meant inserting a blank one, opening its step input and linking it there. It is a category now, like Flow and AI Sandbox, listing the workspace's `ai_agent` resources next to a blank option, filtered by the picker's own search. A picked agent produces a step that is already linked rather than one linked afterwards: `agent` set, no tools, and only the flow-local `user_message`/`user_attachments` transforms. Seeding the brain keys there would leave transforms a linked step never reads and that `AgentResourceBar` strips on its next link change. Each `on:new` forwarder rebuilds the insert detail field by field instead of spreading it, so a new field is dropped unless the forwarder names it. `agentPath` is typed on both `GraphEventHandlers.insert` and `FlowGraphV2`'s `onInsert` so the next one to forget it fails the check. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat: restore the link on cancel and simplify the agent bar Cancel on an agent edit forked the step into a standalone copy, which is the opposite of what the word means and needed a paragraph under the card to explain. It discards the edits and re-links the step now, leaving the agent untouched; diverging from an agent is Unlink's job, on the linked card. This flow's `tool_inputs` survive the round trip as overrides, so Cancel no longer folds them into the tools the way Unlink does. Linking a step to a saved agent happens in the step picker at insert time, so the bar's own resource picker is gone and "Save as agent" is the one action left. Its `+` button was a trap besides: it opened the generic resource form, where an agent would have to be written as raw JSON. The card itself was `surface-secondary`, the sections token, so in dark mode it was darker than the pane and read as a sunken well rather than an elevated card. It uses `surface-tertiary` as the brand table prescribes, its tool chips are `Badge`s, and the editing card no longer overflows the pane and clips its own buttons. The remaining tooltip follows the inline `Label` convention rather than sitting in a flex row whose gap stacked on the trigger's own margin. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat: rework the AI agent evals surface into one table Evals become a single pane: a dataset of cases, one column per scorer, one row per case, with the run being looked at chosen from the toolbar. Runs are permanent. Running the whole dataset opens one; running a single case records nothing at all — it is a job, and looking at what it did is not a claim that it belongs in the history. Its result and its scores sit over the row until they are saved as a run, which carries the cases that were not rerun and the scoring jobs themselves, so the number that is saved is the number that was looked at. A scorer is a runnable: a judge agent or a script, created in one click and edited in place. Scores carry a reason and per-assertion checks, shown on hover with a rescore button. What ran is always named. A run records the agent version, or — for a configuration that is not deployed — a hash of it, so a table can say that its numbers describe an agent that no longer exists: those rows dim and the table offers to rerun. An agent's draft can be run directly instead of the deployed value, and once those edits are deployed the runs that made them are recognised as that version. A step with no agent of its own is evaluable too, and saving it as an agent moves its history onto it. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat: keep an agent's in-progress edits on the agent Editing a linked agent forks it into the step, which is what makes the edits runnable there — but the agent is what is being edited, so that is where the unsaved state belongs. The edit is mirrored into the agent's own resource draft as it is made. It then survives leaving the flow, shows the agent as drafted wherever it appears, and is what evals run when asked to run the draft rather than what is deployed. Deploying or cancelling clears it; opening Edit without changing anything does not mark the agent as drafted. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat: shape the evals surface around a saved agent Evals hang off an `ai_agent` resource, so the surface is now only ever about one: the `draft` subject kind, the standalone-step subject and the move that carried a step's history onto a newly saved agent are gone. - A run is permanent and numbered per agent. Running a single case is a trial: it answers in the panel and never touches the table. - "Run scorers only" opens a run of its own that reuses the answers of the run you are looking at, so a scorer added later measures what already ran without calling the agent again. - A draft run whose configuration is later deployed is stamped, once, to the version it became, so its label stops reading `v23 + edits` forever. - A scorer can carry a pass threshold, read off the scores already recorded. - The table is the case, its answer and one number per scorer; datasets are created and edited in a drawer; a run that executed an earlier state of the current draft says so above the table, in one line. - Which agent a step is, whether it is being edited, and which version it is on is a strip above the step's tabs, because it is true of every tab. - Capturing a case from a step test or a conversation is dropped, and with it the `memory` override on a linked step that nothing set. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat: run past versions of an agent, and number versions per resource The evals home becomes one table of every run of the agent, whichever dataset each is of, with one badge per scorer. A list spanning datasets cannot hold every dataset's scorers to look a name up, so a score carries its name and kind with its number, and thresholds are joined in per run and column. Run now asks what to run: the latest agent, resolved when the run executes as a flow step does, any past version, or the unsaved edits. Pinning is a subject kind of its own, since a linked step resolves the resource live and inlining is the only way to run a version that is no longer current. Scorers move into the edit-dataset drawer. The column header over a run reports and nothing else: a run is permanent, and a control there that changed the columns would edit the past from the one place that must not. Adding one offers four ways rather than two, writing and reusing being different jobs, and both new kinds open with a summary filled in. Versions are numbered per resource. `resource_version.id` is one identity sequence for the whole table, so an agent saved nine times read v4 ... v24, and the gaps counted writes in workspaces the reader cannot see. The id stays how a version is addressed; the new number is what it is called, in the resource history drawer as well as here. It is assigned on write rather than counted on read because trimming past the cap and clearing a history both take the oldest rows, and counting the survivors would renumber a version a run already names. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: read the dataset a remembered selection names Reopening the evals modal restored the last dataset from storage as a bare path, without reading the row it names. Every "is this already the one?" test compared against that selection, so all of them short-circuited and the dataset was never loaded: editing it opened a drawer with no summary, no scorers and no cases. The remembered path is now brought into context the same way any other choice is, and the tests compare against the dataset that is loaded rather than the one that is selected, so a selection can no longer stand for a read that did not happen. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat: give dialogs a trail in their header A dialog deep enough to navigate had nowhere to say where you were: the header held a fixed title, and the way back was a control each body placed for itself, somewhere in a toolbar that moves with everything else the toolbar holds. The header is the one part of the surface that does not move, which is where the trail belongs. `Modal` takes an optional `trail` of levels below its title, rendered as a breadcrumb whose ancestors are the way back. Declarative on purpose: callers of this depth already hold the state that says where they are, so the dialog reads it rather than owning a stack they would have to push and pop in step with it. Escape follows the trail. Leaving a level is what someone deep in a dialog means by it, and closing the whole surface throws away the navigating they did to get there; at the root it closes as before. That only works if a dialog can tell it is the surface being addressed, so `Disposable` now answers `isTopmost()` and the dialog asks before acting: it keeps Escape for itself, so nothing else was arbitrating between it and a drawer opened from inside it, and both were acting on one key press. Evals is the first caller: its runs list is the root, a run is a level in it, and the back button that used to sit above the table is gone. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: portal dialogs out of wherever they were opened from A dialog rendered in place inherits whatever the calling component happens to sit inside. One `transform`, `filter` or `overflow` anywhere above it makes its `fixed` positioning resolve against that ancestor instead of the viewport, and a surface meant to cover the app is then confined to a box it never asked for: the nav rail paints over it and its own edges are clipped. Drawers have always portalled for this reason. Dialogs only did so when an enclosing pane claimed them, and rendered in place otherwise, so the same screen could show a drawer over everything and a dialog trapped behind the nav. They now portal the same way: to the pane when one claims it, to `body` otherwise. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: make the dialog's title the first step of its trail The trail listed levels below the title, so a dialog one level deep read "Evals > All runs > Run 20 · v6": three steps for two places, the first two of them the same place under different names. The title is the root, so it is the root's own segment, and the trail a dialog is given is now the whole path with that segment at its head. Its height stopped moving too. A heading carries a line-height of its own, so a header holding only an h3 stood six pixels shorter than one holding segments as well, and the dialog's whole top edge stepped as you navigated. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: sharpen the evals controls around where you are standing Each screen now offers what belongs to it. The list starts runs; a run is a record, so it offers only the one thing that acts on the record itself, which is measuring the answers it already stored. Starting a fresh run from inside one asked which agent and which dataset from the screen least about either, and scoring an existing run was offered from the list, where there is no run to score. Which run and what it is read against are one question asked twice, so they sit together rather than at opposite ends of a row. Choosing what to run is now a toggle over the two states worth naming, the draft and the saved agent, with every earlier version one click further: running an old version is deliberate, and a list made all three look alike. The draft is read when the dialog opens rather than taken from the caller's polled copy, which could be seconds behind an agent edited a moment ago and would leave the option out exactly when it is the reason for opening the dialog. The dataset field carries its path under it and its edit button on hover, as a resource picker does, so the closed field says what the open list said. Edits waiting on an agent are a "draft" here as everywhere else in Windmill, rather than "+ edits". The dialog runs an evaluation rather than "the agent", which is what it was already called everywhere it is recorded. An agent being edited keeps its evals button on a line of its own, clear of the decision to save or discard. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: settle the evals controls on the patterns Windmill already has The version choice uses ToggleButtonMore, as the AI provider picker does: the two states worth naming stay in the group, the rest are behind the overflow menu, and the one you pick joins the group rather than appearing in a second control below it. The deployed one says which version it resolves to. A run offers nothing to start. Scoring an existing run again was the last thing left there, and it was one button explaining a distinction that the run and the dataset already make between them. The warning that a run executed an earlier draft is about the run on screen, so it goes when the run does rather than following you back to the list, and it sits against the table instead of inside a frame of its own. A dataset just created stays open for its scorers and cases: those are what a dataset is, they can only be added to one that exists, and closing on create sent you to find it again to add them. Scorer settings are a cog rather than a word, now that the row holds three actions. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: close the gap in the version toggle and say what naming a dataset does The overflow trigger is not a pill, so the room it reserves showed as a gap between it and the button before it; it is pulled in by that much. The dataset field gets its clear button, which is also the slot the edit button is positioned against, so the two now sit where a resource picker puts them. Naming a new dataset said nothing about what happens next, and the drawer looked like it was missing the rest of itself. It says so instead: a scorer and a case both belong to a dataset, so there is nothing to attach either to until this one exists, and creating it leaves the drawer open on them. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat: choose a dataset's scorers while naming it A scorer is a reference to a runnable, not a child of the dataset, so it needs the dataset's name but not its row. The list is collected in the drawer while the dataset is being named and sent with the create, which already accepts one, so a dataset arrives holding the columns that were chosen for it rather than being made empty and then edited to hold them. Cases stay where they were: a case *is* a row of the dataset, so there is nothing for it to be a row of until one exists. The drawer says which of the two is which instead of leaving the screen looking like it is missing the rest of itself. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: level the version toggle and name the dataset in its own field The overflow trigger stands a row taller than a toggle button, so the group grew to its height and left the sunken background showing under every pill beside it. Every child of the group is the same height now, which is why the AI provider picker never had the band: it sizes them all alike. The dataset field says the summary with the path after it rather than carrying the path on a line below. The list stacks the two, which a one-line field cannot do, so it says both the other way round. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: tidy the evals forms and the run's own controls Picking a scorer that exists chooses between two sources rather than showing both: the ones already measuring something, and everything else in the workspace. The first list says what each is called with its path under it and what it already measures on the right, instead of three columns that were the same path truncated three ways whenever a scorer had no name of its own. A dataset's drawer says what it is for on the page rather than under an icon, and its summary is sized like the field beneath it. The run's own row lines up with the table under it, the warning above that table is spaced off the rule rather than sitting on it, and adding a case is gone from a run: a run is a record of cases that were answered, so curating them from it is editing what it measured. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat: create a dataset holding the cases written for it Creating a dataset takes the cases to create it with, so one can be assembled in a single act instead of made empty and then filled in. The drawer holds them while the dataset is being named, gives them ids of its own to be edited by, and sends them with the create. Every case is checked before the dataset is written. `eval_case` grants users no write, so the rows cannot be inserted in the transaction that creates the dataset under the caller's own policies; validating first is what keeps "created holding these cases" from becoming "created, holding some of them", and the rows that do follow go in one transaction of their own. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: name the button for what it opens, and say what each version is Starting an evaluation asks which state of the agent and which dataset, and both cost a provider bill, so a button that read as spending one on the way past was lying about the click. It opens something, and says so. Running one case from the panel keeps its own name and its play icon, because that one does run on click. The version options say what they are rather than what they are not: what a flow step would or would not run is a fact about somewhere else, and someone choosing what to evaluate is not standing in it. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: give the editing card two rows and mark evals as beta At the width of a step panel the card's one row wrapped: the line naming the agent, the line saying what saving does, and the two buttons deciding the edits' fate all fought for it. Deciding gets a row of its own, and evals sits against the line it is about, since evals of an agent being edited run the edits. Evals is named wherever it is offered. It read as a word in one state of the card and as an icon in the other, which is two things to recognise for one door. The dialog carries a beta badge against its own name, before any level below it: every way in lands there, so it is said once and stays put as you navigate. The version toggle spells out which is which. Both are the agent at v2 and the difference between them is the whole choice, so it is worth the width. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: name a new dataset, and lay the scorer's settings out like a step's inputs A new dataset arrives called "Dataset 1", which the path follows as it follows any summary: a dataset with none was one every table could only call by its path, and the two seeds are what the summary rule already produces. Scorer settings put each field's description between its label and its input, where a step's inputs put theirs, and its inputs are the size the rest of the drawer uses. The runnable behind the column is a link to it with its kind's icon, since it is a resource of its own and the one thing about it these fields cannot change. The line explaining that a pass line re-reads recorded scores went: the threshold is a number to set, and how it is applied is not a decision being made here. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat: curate a dataset in the drawer and save it in one act The drawer holds the cases while they are edited and writes them when it is saved: added, changed and dropped, whichever it is. Typing no longer writes, so a set is never half saved while someone is still deciding what is in it, and Save means the same thing whether the dataset exists yet or not. A case panel offers reading rather than acting. Running one case now and editing one from a run were the last two ways to change a record from the screen showing it, and the machinery behind the first went with it. The answer is rendered as the prose it is, under what it is: the case's result, whichever run is selected above it. The rest is what the run's table was doing to its own edges: a column name is clipped to its column rather than running into the next, the table squares off against an open panel, and that panel closes with the run it belonged to. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: one border above a table, and a link to the run's job The row above the table drew a bottom border and the table draws its own top edge, so every table sat under two lines. The row keeps its spacing and the table keeps its edge. A column header no longer spins while its scores arrive: the cells under it are where the numbers are missing, and they say so themselves. The beta badge is the height of the word beside it rather than of the line it sits on. A run is one flow and therefore one job, so the run says where that job is: what it is doing, what it cost and what it logged are all there rather than reconstructed from the table. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat: stream scores as each scorer finishes, and show them per case A scorer runs after the agent inside the case's own iteration, so its verdict can be read as soon as its step is done. Waiting for the iteration to end held every column of a case back until the last of them finished, which is why answers arrived one at a time and scores all at once. Reading a job that is still running needs one guard: a module with nothing in it is a step that has not run, not one that produced nothing, and recording the second makes a failure that never goes away. The panel beside the table shows what each column made of the case and why. The reason a judge gave was stored and never shown, which is the half of a score that says anything. It stops repeating the question the header already asks, and a case still running reads as waiting rather than as an answer that says "Running". A run is a number beside a dataset, so the list puts the two together. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat: score a case with every scorer at once The scorers of a case read the answer and never each other, so they ran one after another for no reason: measuring a case now takes as long as its slowest column rather than as long as all of them. Each is a branch of its own, kept from failing the others, so a judge that errors costs its own column and no more. An iteration is three steps again — answer, payload, scores — rather than one per scorer, and each branch is named for the column it produces, so the graph of a run says which scorer did what instead of spelling out an id. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix: read a judge's score out of the JSON it nearly wrote A judge quoting the agent inside its own reason writes those quotes unescaped, which is invalid JSON and also the most ordinary sentence for it to produce. The whole verdict was being thrown away over it, so a column that had a number reported having none. The number and the reason are now read straight out of such text. Deliberately not a second JSON parser: it finds the two keys and takes what follows, which is what survives a quote in the middle of a sentence. A case still running says so with a spinner rather than with the word "Running" sitting where its answer goes. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat: ask a judge for a shape instead of trusting it to write one A new judge carries an output schema, so the provider holds it to `{score, reason}` rather than the prompt asking it to. Windmill already delivers a schema whichever way the model takes it, a tool for Claude and Bedrock and the native parameter elsewhere, so there is no list of models to keep here. An agent with no runs offers its first one where the first row would be, rather than from a toolbar above a table that has nothing in it. Starting a run no longer picks a dataset for you. It fell back to whichever came first, which on an agent that has never run means offering another agent's set as though it were the obvious one; and with no dataset at all it says so and offers the one move there is. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat: report a column that failed throughout, and hold the run dialog The runs overview dropped any column that produced no number, so a judge that failed on every case of a run vanished from the row and read as a column nobody had asked for. The aggregate now reports every column that has cells, with the count of the ones it failed on, and the badge says "failed" where there is nothing to average. A column with no cells at all is still left out: that one was added after the run and has nothing to say about it. Creating a dataset closes the drawer rather than turning it into an edit of what it just made: scorers and cases already ship with the create, so there is nothing left to stay open for. Reached from the run dialog, it gives the screen back with the new dataset selected, and the dialog keeps the version you had already chosen. Also: - the case panel's job link moves to the panel's own header, where its scope is: the job is the whole iteration, not the answer it sat over - one action in the scorer drawer's header, as its neighbours have. The reuse list picks rather than adds, and says which dataset each column already measures - adding a case is the last row of the list it lands in - the pane shows what it has read rather than an empty state it has not earned yet, and its rows say they open - the linked agent card loses a border it had inside another one * fix: keep the linked agent card's outline The card is a thing inside the step's inputs rather than a section of them, and the outline is what says so. Only the rule inside it goes: the detail it separates is already set apart by being detail. * refactor: fit the eval surface to the shipped design * feat: give a nested dialog a back control and the runs list its own moves Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat: put a dialog's description under its title Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * refactor: fold a dialog's back control into the crumb it returns to Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat: edit a dataset's cases as a table rather than a list beside a form Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat: edit a dataset's cases in the grid the data tables are edited in Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat: edit a grid cell of prose in place, and cap a dataset at one page Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * refactor: keep the cell editor's styles beside it, not in the vendored theme Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix: keep an empty cell empty and cap the editor's growth Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat: name the step that assembles a run for the scorers Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat: run the payload step natively, and say so when nothing serves that tag Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix: report an answer as answered while its scorers are still running Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat: let a scorer say a case is not one it measures Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix: score the answer, and leave a case with no expected answer unmeasured Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * refactor: split the evals backend into modules Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix: record what a run produced so it outlives its jobs * fix: read only the agent step's own tool jobs into the payload * fix: pin a run's configuration and give the judge the attachments * feat: write a dataset's cases in one transaction * chore: refresh the sqlx cache for the eval queries * fix: drop results a newer selection has superseded * fix: keep a draft the agent editor never opened on * feat: let a run record what it produced instead of waiting to be read * fix: serialize the replacements of a dataset's cases * fix: stop the poller from superseding a read slower than its interval * chore: refresh the sqlx cache * fix: keep a failed read from settling a cell as a case with no answer * fix: hold the case grid while its save is in flight * fix: keep a failed collect step from failing the run it recorded * chore: refresh the sqlx cache * fix: commit an open cell into the save that reads it * refactor: size the eval buttons with unifiedSize * docs: describe a run as the one flow it is * fix: show a run's recorded rows when part of it cannot be collected * refactor: size the remaining PR-added buttons with unifiedSize * fix: save the dataset name that was submitted, not the one typed after * fix: force an open cell into the save that was pressed for it * fix: refuse to score a run whose evidence could not be read * fix: hold one lock over a dataset's case count and its writes * fix: keep one unreadable run from costing the whole runs list * refactor: drop the banned bindable-default from the eval props * fix: hold the scorer controls while the dataset is written * fix: read only the caller's own draft of an agent * docs: say in the contract that a run pins its configuration * fix: say a scorer did not run rather than blaming a missing answer * feat: resume the agent draft you already had when you press Edit * refactor: build the trail and dataset controls from Button * fix: clear the open-cell flag when the drawer reopens * chore: refresh the sqlx cache * fix: read a run's configuration and its version from one snapshot * fix: refuse a dataset path or summary the column cannot hold * refactor: handle the agent draft the way the resource editor does * fix: run only a configuration the launch actually read * docs: bound dataset path and summary where they are submitted * fix: surface a stalled agent draft instead of claiming it is kept * fix: stop claiming a draft holds edits a failed write never sent * fix: word a missing score only once the run says whether the case answered * fix: let a breadcrumb crumb shrink so its truncation applies * docs: describe where an agent's unsaved edits live and what drops them * fix: keep harvesting scores when the run cannot yet word a missing one * fix: report a refused draft write the card was reading as a save * fix: drop the refused draft write when the server copy is taken instead * refactor: build the scorer and dataset pickers from the design system * fix: say what removing a scorer column actually does * fix: drop a refused draft write wherever the server copy is read * fix: let a picker row be as tall as the two lines it holds * docs: record what removing a scorer column does to recorded runs * fix: send a queued draft write before reopening, and drop only what it refuses * refactor: write the agent draft at commit points instead of mirroring keystrokes Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * refactor: run an agent's edits from the step instead of keeping them as a draft Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix: make the diff badge keyboard operable and refuse an edits run without its edits Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix: drop the dataset icon from the scorer picker rows Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix: size the evals buttons like the rest of windmill and call a run of edits edits Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix: count a brain expression as an edit of the linked agent * fix: cap scorers per dataset and report a launched run as launched * fix: harvest scores in one read, refuse duplicate case ids, allow group paths * fix: mint scorer ids server-side, save a dataset edit in one request, check attachments * fix: write a dataset edit and its cases in one transaction * fix: atomic dataset create/edit, reset eval pane per agent, stable pending scorer ids * refactor: govern eval_case writes by RLS so a dataset edit is one transaction * fix: pin launch snapshot, order case locks, cap dataset size, guard stale load * fix: cap dataset bytes on single-case writes, reset run-dialog flag on load failure * feat: migrate eval datasets on username change, settle unspawned cases, drop unused case endpoints * fix: resolve scorer scripts as the caller and pin their hash; migrate scorer paths on rename * fix: bound a failed tool call's error to the payload truncation cap * fix: pin scorer hash as a hex string, reject missing judges, migrate eval authorship * fix: record an out-of-range scorer result as an error, not a score * fix: resolve judges in one caller-scoped read, pin deployed scripts, bound pass_if * fix: settle unspawned cases only when the run completes, and their score cells too * feat: reassign eval datasets and their path references when offboarding a user * fix: use the regex backreference in offboarding eval path rewrites * fix: register eval datasets in offboarding registries, keep resource-version param name * refactor: name the resource-version path param id, since it is the row id not the version * fix: validate dataset paths canonically, clone eval data on fork, surface eval load and launch failures * docs: note MCP tool results are not yet surfaced to eval scorers * fix: show the eval error state on any load failure, not only an empty dataset list * fix: preserve eval case order across a batched save Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01LVjewUvXFEjNLLw7kxz41h * docs: scope the eval launch delete-safety guarantee to the assembly window Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01LVjewUvXFEjNLLw7kxz41h * fix: only offer deployed scripts as eval scorers, drop unbuilt rescore claim Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01LVjewUvXFEjNLLw7kxz41h * fix: enforce 0-1 scorer threshold in the settings drawer and clear stale eval load errors Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01LVjewUvXFEjNLLw7kxz41h * fix: scope subject version/hash reads to the caller and keep a 0 pass threshold Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01LVjewUvXFEjNLLw7kxz41h * fix: select the saved dataset when creating or renaming from the Run dialog Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01LVjewUvXFEjNLLw7kxz41h * fix: gate eval dataset rename on path ownership, not just write access Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01LVjewUvXFEjNLLw7kxz41h * fix: tolerate a malformed agent config when resolving the deployed label Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01LVjewUvXFEjNLLw7kxz41h * refactor: trim eval code and comments, fix shared select and modal paths Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix: drop the rename warning when editing an eval dataset path Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix: add eval dataset delete, keep summary on partial edits, settle resultless scorer cells Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * test: cover parseThreshold and subjectLabel Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix: hold dataset Save during a scorer write, derive draft_hash only from the carried draft Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Ruben FiszelandClaude Opus 5
|
a9112b72a5 |
fix: make workspace preprocessor scripts selectable in flow preprocessor steps (#10786)
* fix: pick workspace preprocessor scripts in the flow preprocessor step Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01QqKXVsBXynMFtMZ26uvLw7 * fix: explain the empty preprocessor list and keep the editor bar hub populated Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01QqKXVsBXynMFtMZ26uvLw7 * fix: derive the editor bar's script kind from the preprocessor slot Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01QqKXVsBXynMFtMZ26uvLw7 * fix: keep the preprocessor entrypoint when resetting a step's content Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01QqKXVsBXynMFtMZ26uvLw7 * docs: state the preprocessor reset invariant instead of the old control flow Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01QqKXVsBXynMFtMZ26uvLw7 --------- Co-authored-by: Claude Opus 5 <noreply@anthropic.com> |
||
|
hugocasaandClaude Opus 5
|
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> |
||
|
|
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> |
||
|
Ruben Fiszel
|
386c66bef0 | fix: open the expression property column on demand, not from focus (#10558) |