fix(auto-reply,infra): keep startup context and heartbeat event text UTF-16 safe

wings1029 · wings1029 · commit 53892cc55d97 · 2026-07-09T14:50:36.000+08:00
diff --git a/src/auto-reply/reply/startup-context.test.ts b/src/auto-reply/reply/startup-context.test.ts
@@ -5,7 +5,11 @@ import os from "node:os";
 import path from "node:path";
 import { afterEach, describe, expect, it, vi } from "vitest";
 import type { OpenClawConfig } from "../../config/config.js";
-import { buildSessionStartupContextPrelude, shouldApplyStartupContext } from "./startup-context.js";
+import {
+  buildSessionStartupContextPrelude,
+  shouldApplyStartupContext,
+  trimStartupMemoryContent,
+} from "./startup-context.js";
 
 const tmpDirs: string[] = [];
 
@@ -493,3 +497,22 @@ describe("shouldApplyStartupContext", () => {
     expect(shouldApplyStartupContext({ cfg: applyOnCfg, action: "reset" })).toBe(false);
   });
 });
+
+describe("trimStartupMemoryContent", () => {
+  it("returns the original text when it fits within maxChars", () => {
+    expect(trimStartupMemoryContent("hello", 100)).toBe("hello");
+  });
+
+  it("trims whitespace and truncates long content", () => {
+    const input = `  ${"x".repeat(200)}  `;
+    const result = trimStartupMemoryContent(input, 100);
+    expect(result.length).toBeLessThan(input.length);
+    expect(result).toContain("[truncated]");
+  });
+
+  it("does not split a surrogate pair at the boundary", () => {
+    const input = `aa🚀${"b".repeat(200)}`;
+    const result = trimStartupMemoryContent(input, 80);
+    expect(result).not.toContain("�");
+  });
+});
diff --git a/src/auto-reply/reply/startup-context.ts b/src/auto-reply/reply/startup-context.ts
@@ -3,6 +3,7 @@ import fs from "node:fs";
 import path from "node:path";
 import { resolveIntegerOption } from "@openclaw/normalization-core/number-coercion";
 import { uniqueStrings } from "@openclaw/normalization-core/string-normalization";
+import { truncateUtf16Safe } from "@openclaw/normalization-core/utf16-slice";
 import { formatDateStamp, resolveUserTimezone } from "../../agents/date-time.js";
 import type { OpenClawConfig } from "../../config/config.js";
 import { openRootFile } from "../../infra/boundary-file-read.js";
@@ -89,12 +90,12 @@ function buildStartupMemoryDateStamps(params: {
     : [...localWindow, utcTodayStamp];
 }
 
-function trimStartupMemoryContent(content: string, maxChars: number): string {
+export function trimStartupMemoryContent(content: string, maxChars: number): string {
   const trimmed = content.trim();
   if (trimmed.length <= maxChars) {
     return trimmed;
   }
-  return `${trimmed.slice(0, maxChars)}\n...[truncated]...`;
+  return `${truncateUtf16Safe(trimmed, maxChars)}\n...[truncated]...`;
 }
 
 function escapeQuotedStartupMemory(content: string): string {
diff --git a/src/infra/heartbeat-events-filter.test.ts b/src/infra/heartbeat-events-filter.test.ts
@@ -211,3 +211,18 @@ describe("isExecCompletionEvent", () => {
     expect(isExecCompletionEvent("Check: exec completed (last run was yesterday)")).toBe(false);
   });
 });
+
+describe("buildExecEventPrompt truncation", () => {
+  it("does not split surrogate pairs in long event text", () => {
+    const emojiLine = `result: aa🚀${"b".repeat(8_000)}`;
+    const result = buildExecEventPrompt([emojiLine]);
+    expect(result).not.toContain("�");
+    expect(result.length).toBeLessThanOrEqual(8_500); // body + notice
+  });
+
+  it("passes through short event text unchanged", () => {
+    const result = buildExecEventPrompt(["hello"]);
+    expect(result).toContain("hello");
+    expect(result).not.toContain("[truncated]");
+  });
+});
diff --git a/src/infra/heartbeat-events-filter.ts b/src/infra/heartbeat-events-filter.ts
@@ -1,5 +1,6 @@
 // Filters heartbeat event text before it is added to prompts.
 import { normalizeLowercaseStringOrEmpty } from "@openclaw/normalization-core/string-coerce";
+import { truncateUtf16Safe } from "@openclaw/normalization-core/utf16-slice";
 import { HEARTBEAT_RESPONSE_TOOL_INSTRUCTIONS } from "../auto-reply/heartbeat.js";
 import { HEARTBEAT_TOKEN } from "../auto-reply/tokens.js";
 
@@ -124,7 +125,7 @@ export function buildExecEventPrompt(
   const { text: rawEventText, hasMissingOutputFailure } = formatExecEventPromptText(pendingEvents);
   const eventText =
     rawEventText.length > MAX_EXEC_EVENT_PROMPT_CHARS
-      ? `${rawEventText.slice(0, MAX_EXEC_EVENT_PROMPT_CHARS)}\n\n[truncated]`
+      ? `${truncateUtf16Safe(rawEventText, MAX_EXEC_EVENT_PROMPT_CHARS)}\n\n[truncated]`
       : rawEventText;
   if (!eventText) {
     if (useHeartbeatResponseTool) {
