diff --git a/extensions/whatsapp/src/send.test.ts b/extensions/whatsapp/src/send.test.ts
index bf17b4131b7a..e4a51122d740 100644
--- a/extensions/whatsapp/src/send.test.ts
+++ b/extensions/whatsapp/src/send.test.ts
@@ -4,6 +4,7 @@ import os from "node:os";
 import path from "node:path";
 import type { OpenClawConfig } from "openclaw/plugin-sdk/config-contracts";
 import { redactIdentifier } from "openclaw/plugin-sdk/logging-core";
+import { MEDIA_FFMPEG_MAX_AUDIO_DURATION_SECS } from "openclaw/plugin-sdk/media-runtime";
 import { afterEach, beforeAll, beforeEach, describe, expect, it, vi } from "vitest";
 import type { WhatsAppSendKind, WhatsAppSendResult } from "./inbound/send-result.js";
 import type { ActiveWebListener } from "./inbound/types.js";
@@ -296,9 +297,28 @@ describe("web outbound", () => {
       mediaUrl: `/tmp/${media.fileName}`,
     });
 
-    expect(hoisted.runFfmpeg).toHaveBeenCalledWith(
-      expect.arrayContaining(["-c:a", "libopus", "-ar", "48000", "-b:a", "64k"]),
-    );
+    expect(hoisted.runFfmpeg).toHaveBeenCalledTimes(1);
+    const ffmpegArgs = hoisted.runFfmpeg.mock.calls[0]?.[0] as string[] | undefined;
+    expect(ffmpegArgs?.slice(0, 5)).toEqual(["-hide_banner", "-loglevel", "error", "-y", "-i"]);
+    expect(ffmpegArgs?.[5]).toContain(`/input.${media.name}`);
+    expect(ffmpegArgs?.slice(6, -1)).toEqual([
+      "-vn",
+      "-sn",
+      "-dn",
+      "-t",
+      String(MEDIA_FFMPEG_MAX_AUDIO_DURATION_SECS),
+      "-ar",
+      "48000",
+      "-ac",
+      "1",
+      "-c:a",
+      "libopus",
+      "-b:a",
+      "64k",
+    ]);
+    const outputPath = ffmpegArgs?.at(-1);
+    expect(outputPath).toContain("/fs-safe-output-");
+    expect(outputPath).toContain("-voice.ogg.part");
     expect(sendMessage).toHaveBeenNthCalledWith(
       1,
       "+1555",
@@ -381,12 +401,12 @@ describe("web outbound", () => {
       mediaUrls: [" /tmp/secondary.jpg "],
     });
 
-    expect(loadWebMediaMock).toHaveBeenCalledWith(
-      "/tmp/primary.jpg",
-      expect.objectContaining({
-        hostReadCapability: false,
-      }),
-    );
+    expect(loadWebMediaMock).toHaveBeenCalledWith("/tmp/primary.jpg", {
+      maxBytes: 50 * 1024 * 1024,
+      localRoots: undefined,
+      readFile: undefined,
+      hostReadCapability: false,
+    });
     expect(sendMessage).toHaveBeenLastCalledWith("+1555", "pic", buf, "image/jpeg");
   });
 
@@ -402,12 +422,12 @@ describe("web outbound", () => {
       cfg: WHATSAPP_TEST_CFG,
       mediaUrls: ["   ", " /tmp/pic.jpg "],
     });
-    expect(loadWebMediaMock).toHaveBeenCalledWith(
-      "/tmp/pic.jpg",
-      expect.objectContaining({
-        hostReadCapability: false,
-      }),
-    );
+    expect(loadWebMediaMock).toHaveBeenCalledWith("/tmp/pic.jpg", {
+      maxBytes: 50 * 1024 * 1024,
+      localRoots: undefined,
+      readFile: undefined,
+      hostReadCapability: false,
+    });
     expect(sendMessage).toHaveBeenLastCalledWith("+1555", "pic", buf, "image/jpeg");
   });
 
@@ -463,13 +483,12 @@ describe("web outbound", () => {
       mediaLocalRoots: ["/tmp/workspace"],
     });
 
-    expect(loadWebMediaMock).toHaveBeenCalledWith(
-      "/tmp/pic.jpg",
-      expect.objectContaining({
-        maxBytes: 100 * 1024 * 1024,
-        localRoots: ["/tmp/workspace"],
-      }),
-    );
+    expect(loadWebMediaMock).toHaveBeenCalledWith("/tmp/pic.jpg", {
+      maxBytes: 100 * 1024 * 1024,
+      localRoots: ["/tmp/workspace"],
+      readFile: undefined,
+      hostReadCapability: false,
+    });
   });
 
   it("sends polls via active listener", async () => {