fix: the cron isolated agent in openclaw unconditiona (#383) (#63878)

CHANGELOG.md

@@ -119,6 +119,7 @@ Docs: https://docs.openclaw.ai
 - Agents/exec: keep sandboxed `tools.exec.host=auto` sessions from honoring per-call `host=node` or `host=gateway` overrides while a sandbox runtime is active, and stop advertising node routing in that state so exec stays on the sandbox host. (#63880)
 - Gateway/restart sentinel: route restart notices only from stored canonical delivery metadata and skip outbound guessing from lossy session keys, avoiding misdelivery on case-sensitive channels like Matrix. (#64391) Thanks @gumadeiras.
 
+- Cron/isolated agent: run scheduled agent turns as non-owner senders so owner-only tools stay unavailable during cron execution. (#63878)
 ## 2026.4.9
 
 ### Changes

src/cron/isolated-agent/run-executor.ts

@@ -131,7 +131,7 @@ export function createCronPromptExecutor(params: {
           agentId: params.agentId,
           trigger: "cron",
           allowGatewaySubagentBinding: true,
-          senderIsOwner: true,
+          senderIsOwner: false,
           messageChannel: params.messageChannel,
           agentAccountId: params.resolvedDelivery.accountId,
           sessionFile,

src/cron/isolated-agent/run.owner-auth.test.ts

@@ -58,14 +58,14 @@ describe("runCronIsolatedAgentTurn owner auth", () => {
   });
 
   it(
-    "passes senderIsOwner=true to isolated cron agent runs",
+    "passes senderIsOwner=false to isolated cron agent runs",
     { timeout: RUN_OWNER_AUTH_TIMEOUT_MS },
     async () => {
       await runCronIsolatedAgentTurn(makeParams());
 
       expect(runEmbeddedPiAgentMock).toHaveBeenCalledTimes(1);
       const senderIsOwner = runEmbeddedPiAgentMock.mock.calls[0]?.[0]?.senderIsOwner;
-      expect(senderIsOwner).toBe(true);
+      expect(senderIsOwner).toBe(false);
     },
   );
 });