fix: remove hardcoded disableBlockStreaming to honor agent config for TUI (#19693)

Merged via /review-pr -> /prepare-pr -> /merge-pr.

Prepared head SHA: 710d449080
Co-authored-by: neipor <191749196+neipor@users.noreply.github.com>
Co-authored-by: gumadeiras <5599352+gumadeiras@users.noreply.github.com>
Reviewed-by: @gumadeiras

CHANGELOG.md

@@ -16,6 +16,7 @@ Docs: https://docs.openclaw.ai
 ### Fixes
 
 - iOS/Screen: move `WKWebView` lifecycle ownership into `ScreenWebView` coordinator and explicit attach/detach flow to reduce gesture/lifecycle crash risk (`__NSArrayM insertObject:atIndex:` paths) during screen tab updates. (#20366) Thanks @ngutman.
+- Gateway/TUI: honor `agents.defaults.blockStreamingDefault` for `chat.send` by removing the hardcoded block-streaming disable override, so replies can use configured block-mode delivery. (#19693) Thanks @neipor.
 - Protocol/Apple: regenerate Swift gateway models for `push.test` so `pnpm protocol:check` stays green on main. Thanks @mbelinky.
 - Canvas/A2UI: improve bundled-asset resolution and empty-state handling so UI fallbacks render reliably. (#20312) Thanks @mbelinky.
 - UI/Sessions: accept the canonical main session-key alias in Chat UI flows so main-session routing stays consistent. (#20311) Thanks @mbelinky.

src/gateway/server-methods/chat.ts

@@ -898,7 +898,6 @@ export const chatHandlers: GatewayRequestHandlers = {
           runId: clientRunId,
           abortSignal: abortController.signal,
           images: parsedImages.length > 0 ? parsedImages : undefined,
-          disableBlockStreaming: true,
           onAgentRunStart: (runId) => {
             agentRunStarted = true;
             const connId = typeof client?.connId === "string" ? client.connId : undefined;

src/gateway/server.chat.gateway-server-chat-b.e2e.test.ts

@@ -2,6 +2,7 @@ import fs from "node:fs/promises";
 import os from "node:os";
 import path from "node:path";
 import { describe, expect, test, vi } from "vitest";
+import type { GetReplyOptions } from "../auto-reply/types.js";
 import { __setMaxChatHistoryMessagesBytesForTest } from "./server-constants.js";
 import {
   connectOk,
@@ -136,6 +137,42 @@ describe("gateway server chat", () => {
     });
   });
 
+  test("chat.send does not force-disable block streaming", async () => {
+    await withGatewayChatHarness(async ({ ws, createSessionDir }) => {
+      const spy = getReplyFromConfig;
+      await connectOk(ws);
+
+      await createSessionDir();
+      await writeMainSessionStore();
+      testState.agentConfig = { blockStreamingDefault: "on" };
+      try {
+        spy.mockReset();
+        let capturedOpts: GetReplyOptions | undefined;
+        spy.mockImplementationOnce(async (_ctx: unknown, opts?: GetReplyOptions) => {
+          capturedOpts = opts;
+        });
+
+        const sendRes = await rpcReq(ws, "chat.send", {
+          sessionKey: "main",
+          message: "hello",
+          idempotencyKey: "idem-block-streaming",
+        });
+        expect(sendRes.ok).toBe(true);
+
+        await vi.waitFor(
+          () => {
+            expect(spy.mock.calls.length).toBeGreaterThan(0);
+          },
+          { timeout: 2_000, interval: 10 },
+        );
+
+        expect(capturedOpts?.disableBlockStreaming).toBeUndefined();
+      } finally {
+        testState.agentConfig = undefined;
+      }
+    });
+  });
+
   test("chat.history hard-caps single oversized nested payloads", async () => {
     await withGatewayChatHarness(async ({ ws, createSessionDir }) => {
       const historyMaxBytes = 64 * 1024;
@@ -251,7 +288,7 @@ describe("gateway server chat", () => {
 
   test("smoke: supports abort and idempotent completion", async () => {
     await withGatewayChatHarness(async ({ ws, createSessionDir }) => {
-      const spy = vi.mocked(getReplyFromConfig) as unknown as ReturnType<typeof vi.fn>;
+      const spy = getReplyFromConfig;
       let aborted = false;
       await connectOk(ws);
 

src/gateway/test-helpers.mocks.ts

@@ -4,7 +4,10 @@ import fs from "node:fs/promises";
 import os from "node:os";
 import path from "node:path";
 import { Mock, vi } from "vitest";
+import type { MsgContext } from "../auto-reply/templating.js";
+import type { GetReplyOptions, ReplyPayload } from "../auto-reply/types.js";
 import type { ChannelPlugin, ChannelOutboundAdapter } from "../channels/plugins/types.js";
+import type { OpenClawConfig } from "../config/config.js";
 import { applyPluginAutoEnable } from "../config/plugin-auto-enable.js";
 import type { AgentBinding } from "../config/types.agents.js";
 import type { HooksConfig } from "../config/types.hooks.js";
@@ -19,6 +22,12 @@ type StubChannelOptions = {
   summary?: Record<string, unknown>;
 };
 
+type GetReplyFromConfigFn = (
+  ctx: MsgContext,
+  opts?: GetReplyOptions,
+  configOverride?: OpenClawConfig,
+) => Promise<ReplyPayload | ReplyPayload[] | undefined>;
+
 const createStubOutboundAdapter = (channelId: ChannelPlugin["id"]): ChannelOutboundAdapter => ({
   deliveryMode: "direct",
   sendText: async () => ({
@@ -169,7 +178,7 @@ const hoisted = vi.hoisted(() => ({
     waitResults: new Map<string, boolean>(),
   },
   testTailscaleWhois: { value: null as TailscaleWhoisIdentity | null },
-  getReplyFromConfig: vi.fn().mockResolvedValue(undefined),
+  getReplyFromConfig: vi.fn<GetReplyFromConfigFn>().mockResolvedValue(undefined),
   sendWhatsAppMock: vi.fn().mockResolvedValue({ messageId: "msg-1", toJid: "jid-1" }),
 }));
 
@@ -202,7 +211,7 @@ export const testTailscaleWhois = hoisted.testTailscaleWhois;
 export const piSdkMock = hoisted.piSdkMock;
 export const cronIsolatedRun = hoisted.cronIsolatedRun;
 export const agentCommand: Mock<() => void> = hoisted.agentCommand;
-export const getReplyFromConfig: Mock<() => void> = hoisted.getReplyFromConfig;
+export const getReplyFromConfig: Mock<GetReplyFromConfigFn> = hoisted.getReplyFromConfig;
 
 export const testState = {
   agentConfig: undefined as Record<string, unknown> | undefined,