Merge branch 'main' into feat/734-add-apify-to-the-recommended-providers-list

Author: daveomri

.changeset/cost-skill-webhook-filters.md

@@ -0,0 +1,5 @@
+---
+"@trigger.dev/sdk": patch
+---
+
+The bundled `trigger-cost-savings` skill now notes that a webhook delivery starts no run when no subscriber's filter matches, so it costs no compute.

.changeset/head-start-parallel-tool-results.md

@@ -0,0 +1,5 @@
+---
+"@trigger.dev/sdk": patch
+---
+
+Fix a `chat.agent` turn failing after a head start that ended with two or more parallel tool calls. When those tools finished out of call order, the agent sent their results to the model twice on the next step, and the provider rejected the request (Anthropic: "each tool_use must have a single result"). Each handed-over tool result is now sent once.

.changeset/socket-io-client-ws.md

@@ -0,0 +1,6 @@
+---
+"@trigger.dev/core": patch
+"trigger.dev": patch
+---
+
+Upgrade `socket.io-client` to 4.8.4 so projects using the SDK and CLI install `ws` 8.21 instead of 8.17.1, which has published advisories (GHSA-96hv-2xvq-fx4p, GHSA-58qx-3vcg-4xpx). The old pin blocked upgrading `ws` without a package manager override.

.server-changes/batch-results-sharded-batch-ids.md

@@ -0,0 +1,6 @@
+---
+area: webapp
+type: fix
+---
+
+Fix the batch results API returning "Batch not found" for some batches that the batch retrieve API could see.

.server-changes/child-run-trace-api-faster.md

@@ -0,0 +1,6 @@
+---
+area: webapp
+type: fix
+---
+
+Fix fetching the trace of a child run through the API sometimes being very slow or failing. These requests now return quickly.

.server-changes/email-change-confirmation.md

@@ -0,0 +1,6 @@
+---
+area: webapp
+type: fix
+---
+
+Changing your email address now requires clicking a confirmation button after opening the emailed link.

.server-changes/route-v1-event-store-to-v2.md

@@ -0,0 +1,6 @@
+---
+area: webapp
+type: breaking
+---
+
+Self-hosted instances on the legacy `clickhouse` event store now use `clickhouse_v2`, so run logs and traces recorded before the upgrade no longer show. Set `EVENT_REPOSITORY_DEFAULT_STORE=clickhouse_v2`, or set `EVENT_REPOSITORY_V1_REROUTE_ENABLED=0` to stay on the legacy store.

.server-changes/run-snapshot-ordering.md

@@ -0,0 +1,6 @@
+---
+area: webapp
+type: fix
+---
+
+Fix a rare case where clock skew between run engine instances could leave a run executing after its task finished, or stop a wait from ever resuming. A run's state changes are now always ordered in the order they happened, regardless of clock skew.

apps/supervisor/package.json

@@ -29,6 +29,6 @@
   "devDependencies": {
     "@internal/testcontainers": "workspace:*",
     "@types/dockerode": "^3.3.33",
-    "socket.io-client": "4.7.5"
+    "socket.io-client": "4.8.4"
   }
 }

apps/webapp/app/components/runs/v3/trace/useRunTrace.ts

@@ -71,20 +71,25 @@ export function useRunTrace({
   errorsOnly,
   maximumLiveReloadingSetting,
   onRunDataStale,
+  readContext,
 }: {
   run: RunTraceRun;
   trace: RunTraceInput;
   errorsOnly: boolean;
   maximumLiveReloadingSetting: number;
   onRunDataStale: () => void;
+  /** Sent with every chunk read; "session" bounds completed runs' reads. */
+  readContext?: "session";
 }) {
   const organization = useOrganization();
   const project = useProject();
   const environment = useEnvironment();
 
   const { isTruncated = false, missingAnchor = false } = trace;
 
-  const chunkPath = v3RunTraceChunkPath(organization, project, environment, run);
+  const chunkPath = v3RunTraceChunkPath(organization, project, environment, run, {
+    context: readContext,
+  });
   const statusPath = v3RunStatusPath(organization, project, environment, run);
   const {
     events,