aimeat 1.19.0 → 1.19.1

package/dist/locales/en.json

@@ -9,6 +9,10 @@
   "profile.scheduler.newSchedule": "New schedule",
   "profile.scheduler.close": "Close",
   "profile.scheduler.triggered": "Schedule run started",
+  "profile.scheduler.runCreated": "Run started — a new task was created",
+  "profile.scheduler.runBusy": "No task created — a previous run is still active. Finish, fail, or delete it, then run again.",
+  "profile.scheduler.runLimited": "Not run — this schedule's run limit (daily or max) has been reached.",
+  "profile.scheduler.runError": "Run failed",
   "profile.scheduler.cancelled": "Schedule cancelled",
   "profile.scheduler.created": "Schedule created",
   "profile.scheduler.saving": "Saving…",

package/dist/locales/fi.json

@@ -9,6 +9,10 @@
   "profile.scheduler.newSchedule": "Uusi ajastus",
   "profile.scheduler.close": "Sulje",
   "profile.scheduler.triggered": "Ajastuksen suoritus käynnistetty",
+  "profile.scheduler.runCreated": "Suoritus aloitettu — uusi tehtävä luotiin",
+  "profile.scheduler.runBusy": "Tehtävää ei luotu — edellinen suoritus on yhä käynnissä. Päätä, merkitse epäonnistuneeksi tai poista se, ja suorita uudelleen.",
+  "profile.scheduler.runLimited": "Ei suoritettu — tämän ajastuksen suoritusraja (päivittäinen tai maksimi) on saavutettu.",
+  "profile.scheduler.runError": "Suoritus epäonnistui",
   "profile.scheduler.cancelled": "Ajastus peruttu",
   "profile.scheduler.created": "Ajastus luotu",
   "profile.scheduler.saving": "Tallennetaan…",

package/dist/public/views/profile/agents-tasks-subtab.js

@@ -9,6 +9,11 @@
  *   - TaskItem -- task row with expand/collapse, todo list, start button
  *   - RequestChangesModal -- inline modal for owner to send a free-text change request
  * @version-history
+ *   v4.11.0 -- 2026-06-05 -- Show the Delete button on every non-active task
+ *     (was queued/draft/revision_requested only) so done/failed/paused/stalled/
+ *     archived tasks can be removed; active tasks still use Cancel. The backend
+ *     delete also cleans the task's operational traces (events, live-status
+ *     memory keys, cancel marker), preserving the deliverable.
  *   v4.10.0 -- 2026-06-03 -- Deep-link support: AgentTasksSubtab accepts
  *     openTaskId + openTaskNonce; a nonce bump resets the bucket to 'recent'
  *     (clears search/time) so the target task is visible, and TaskItem auto-opens
@@ -581,6 +586,11 @@ function TaskItem({ task, agentName, showToast, onRefresh, autoOpen = 0 }) {
   const isRevisionRequested = task.status === 'revision_requested';
   const isActive = task.status === 'active';
   const isDone = task.status === 'done';
+  // Any non-active task can be deleted (done/failed/paused/stalled/queued/draft/
+  // revision_requested). An active task uses Cancel instead -- the owner stops
+  // the running work first, then deletes it. Delete also cleans the task's
+  // operational traces server-side (event log, live-status keys, cancel marker).
+  const canDelete = !isActive;
   const rating = task.rating;
   const canStart = isQueued && hasTodos;
   const canRequestChanges = task.status === 'queued' && hasTodos;
@@ -787,7 +797,7 @@ function TaskItem({ task, agentName, showToast, onRefresh, autoOpen = 0 }) {
             ${(isActive || task.status === 'stalled') && html`
               <button class="btn-danger btn-sm" onClick=${handleCancel}>${t('profile.agents.tasks.cancel')}</button>
             `}
-            ${(isQueued || task.status === 'draft' || isRevisionRequested) && html`
+            ${canDelete && html`
               <button class="btn-danger btn-sm" onClick=${handleDelete}>${t('profile.agents.tasks.delete')}</button>
             `}
             <span class="pf-agd-task-actions-spacer"></span>

package/dist/public/views/profile/schedule-item.js

@@ -9,6 +9,9 @@
  *   edit (PATCH), and cancel itself; calls onChanged() to let the parent refetch.
  * @version-history
  *   v1.0.0 -- 2026-06-03 -- Initial editable schedule card
+ *   v1.1.0 -- 2026-06-05 -- "Run now" reads the trigger outcome and reports it:
+ *     a warning toast when no task was created (a previous run is still active,
+ *     or a run limit was reached) instead of a misleading "started" success.
  */
 import { h } from 'preact';
 import { useState } from 'preact/hooks';
@@ -82,7 +85,18 @@ export default function ScheduleItem({ schedule: s, onChanged, showToast }) {
     try { await fn(); onChanged?.(); } catch (e) { showToast?.(e.message, true); } finally { setBusy(false); }
   };
   const onToggle = () => run(() => setScheduleEnabled(s.id, !s.enabled));
-  const onTrigger = () => run(async () => { await triggerSchedule(s.id); showToast?.(t('profile.scheduler.triggered')); });
+  // "Run now" surfaces what actually happened: agent_task schedules can decline
+  // to create a task (a previous run is still active, or a run limit is reached),
+  // which used to look like success. Map the backend outcome to a clear toast.
+  const onTrigger = () => run(async () => {
+    const res = await triggerSchedule(s.id);
+    const d = res?.data || {};
+    if (d.outcome === 'created') showToast?.(t('profile.scheduler.runCreated'));
+    else if (d.outcome === 'busy') showToast?.(t('profile.scheduler.runBusy'), true);
+    else if (d.outcome === 'limited') showToast?.(t('profile.scheduler.runLimited'), true);
+    else if (d.outcome === 'error') showToast?.(d.reason || t('profile.scheduler.runError'), true);
+    else showToast?.(t('profile.scheduler.triggered')); // 'ran' (ai/extension) or older server
+  });
   const onCancel = () => { if (window.confirm(t('profile.scheduler.confirmCancel'))) run(() => deleteSchedule(s.id)); };
 
   const onSave = async () => {

package/dist/src/generated/api-types.d.ts

@@ -7141,7 +7141,10 @@ export interface paths {
         };
         get?: never;
         put?: never;
-        /** Run a schedule immediately */
+        /**
+         * Run a schedule immediately
+         * @description Runs the schedule now (owner only). The response `data.outcome` reports what happened: `created` (an agent_task occurrence was queued/activated — `data.task_id` is set), `ran` (a non-task ai/extension job executed), `busy` (skipped — a previous occurrence is still running; `data.reason` explains), `limited` (skipped — a run limit such as daily/max was reached), or `error` (the job ran but failed; `data.reason` has the message). A manual run only defers to a genuinely running (active/stalled) occurrence — a paused/queued/archived one no longer blocks it.
+         */
         post: operations["triggerSchedule"];
         delete?: never;
         options?: never;
@@ -7227,7 +7230,7 @@ export interface paths {
         post?: never;
         /**
          * Delete a task
-         * @description Delete a task. Only draft and queued tasks can be deleted.
+         * @description Delete a task (owner only). Any task can be deleted EXCEPT one that is currently `active` -- a running task must be cancelled or paused first. Deletion also removes the task's event log and cleans its operational memory traces (the agent's live-status keys `agents.{name}.tasks.{id}.*` and the owner-written cancel marker `agents.cancel.task.{id}`). The agent's produced deliverable/output memory is preserved.
          */
         delete: operations["deleteAgentTask"];
         options?: never;
@@ -21521,7 +21524,7 @@ export interface operations {
         };
         requestBody?: never;
         responses: {
-            /** @description Triggered */
+            /** @description Triggered. `data.outcome` is one of created | ran | busy | limited | error; `data.task_id` is present when a task was created; `data.reason` is present on busy/limited/error. */
             200: {
                 headers: {
                     [name: string]: unknown;
@@ -21757,6 +21760,7 @@ export interface operations {
             401: components["responses"]["Unauthorized"];
             403: components["responses"]["Forbidden"];
             404: components["responses"]["NotFound"];
+            409: components["responses"]["Conflict"];
         };
     };
     updateAgentTask: {