fix(agents): skip redundant partial compaction summarization (#61603) (thanks @neeravmakwana)

neeravmakwana · web-flow · commit 7df5f7024266 · 2026-04-06T13:21:07.000+01:00
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -250,6 +250,7 @@ Docs: https://docs.openclaw.ai
 - Plugins: suppress trust-warning noise during non-activating snapshot and CLI metadata loads. (#61427) Thanks @gumadeiras.
 - Agents/video generation: accept `agents.defaults.videoGenerationModel` in strict config validation and `openclaw config set/get`, so gateways using `video_generate` no longer fail to boot after enabling a video model.
 - Matrix/streaming: add a quiet preview mode for streamed Matrix replies, keep legacy `partial` preview-first behavior, and finalize quiet media captions correctly so previews stop notifying early without dropping final text semantics. (#61450) Thanks @gumadeiras.
+- Agents/compaction: skip redundant partial summarization when no messages were oversized, so the same transcript is not summarized twice after a full summarization failure. Fixes #61465. (#61603) Thanks @neeravmakwana.
 - Gateway/shutdown: bound websocket-server shutdown even when no tracked clients remain, so gateway restarts stop hanging until the watchdog kills the process. (#61565) Thanks @mbelinky.
 - Control UI/multilingual: localize the remaining shared channel, instances, nodes, and gateway-confirmation strings so the dashboard stops mixing translated UI with hardcoded English labels. Thanks @vincentkoc.
 - Discord/media: raise the default inbound and outbound media cap to `100MB` so Discord matches Telegram more closely and larger attachments stop failing on the old low default.
diff --git a/src/agents/compaction.summarize-fallback.test.ts b/src/agents/compaction.summarize-fallback.test.ts
@@ -0,0 +1,103 @@
+import type { AgentMessage } from "@mariozechner/pi-agent-core";
+import type { UserMessage } from "@mariozechner/pi-ai";
+import type { ExtensionContext } from "@mariozechner/pi-coding-agent";
+import { beforeEach, describe, expect, it, vi } from "vitest";
+
+const piCodingAgentMocks = vi.hoisted(() => ({
+  generateSummary: vi.fn(),
+  estimateTokens: vi.fn((_message: unknown) => 100),
+}));
+
+vi.mock("@mariozechner/pi-coding-agent", async () => {
+  const actual = await vi.importActual<typeof import("@mariozechner/pi-coding-agent")>(
+    "@mariozechner/pi-coding-agent",
+  );
+  return {
+    ...actual,
+    generateSummary: piCodingAgentMocks.generateSummary,
+    estimateTokens: piCodingAgentMocks.estimateTokens,
+  };
+});
+
+const { summarizeWithFallback } = await import("./compaction.js");
+
+const testModel = {
+  id: "test",
+  name: "test",
+  contextWindow: 200_000,
+  contextTokens: 200_000,
+  maxTokens: 8192,
+} as unknown as NonNullable<ExtensionContext["model"]>;
+
+describe("summarizeWithFallback", () => {
+  beforeEach(() => {
+    piCodingAgentMocks.generateSummary.mockReset();
+    piCodingAgentMocks.generateSummary.mockRejectedValue(
+      new Error("Summarization failed: fetch failed"),
+    );
+    piCodingAgentMocks.estimateTokens.mockReset();
+    piCodingAgentMocks.estimateTokens.mockImplementation(() => 100);
+  });
+
+  it("does not duplicate summarization when no messages were oversized", async () => {
+    const messages: AgentMessage[] = [
+      {
+        role: "user",
+        content: "hello",
+        timestamp: 1,
+      } satisfies UserMessage,
+    ];
+
+    const result = await summarizeWithFallback({
+      messages,
+      model: testModel,
+      apiKey: "test-key", // pragma: allowlist secret
+      signal: new AbortController().signal,
+      reserveTokens: 1000,
+      maxChunkTokens: 50_000,
+      contextWindow: 200_000,
+    });
+
+    expect(result).toContain("Context contained 1 messages");
+    expect(result).toContain("0 oversized");
+    // Full path: retryAsync attempts (3) for a single chunk; partial path must not run.
+    expect(piCodingAgentMocks.generateSummary).toHaveBeenCalledTimes(3);
+  });
+
+  it("still attempts partial summarization when oversized messages were excluded", async () => {
+    piCodingAgentMocks.estimateTokens.mockImplementation((message: unknown) => {
+      const content =
+        typeof (message as { content?: unknown }).content === "string"
+          ? (message as { content: string }).content
+          : "";
+      return content.length > 10_000 ? 500_000 : 100;
+    });
+
+    const messages: AgentMessage[] = [
+      {
+        role: "user",
+        content: "small",
+        timestamp: 1,
+      } satisfies UserMessage,
+      {
+        role: "user",
+        content: "x".repeat(500_000),
+        timestamp: 2,
+      } satisfies UserMessage,
+    ];
+
+    const result = await summarizeWithFallback({
+      messages,
+      model: testModel,
+      apiKey: "test-key", // pragma: allowlist secret
+      signal: new AbortController().signal,
+      reserveTokens: 1000,
+      maxChunkTokens: 50_000,
+      contextWindow: 200_000,
+    });
+
+    expect(result).toContain("2 messages (1 oversized)");
+    // Full attempt (3 retries) plus distinct partial transcript (3 retries).
+    expect(piCodingAgentMocks.generateSummary.mock.calls.length).toBe(6);
+  });
+});
diff --git a/src/agents/compaction.ts b/src/agents/compaction.ts
@@ -398,7 +398,7 @@ export async function summarizeWithFallback(params: {
     return await summarizeChunks(params);
   } catch (fullError) {
     log.warn(
-      `Full summarization failed, trying partial: ${
+      `Full summarization failed: ${
         fullError instanceof Error ? fullError.message : String(fullError)
       }`,
     );
@@ -420,7 +420,9 @@ export async function summarizeWithFallback(params: {
     }
   }
 
-  if (smallMessages.length > 0) {
+  // When nothing was oversized, `smallMessages` is the same transcript as the full attempt.
+  // Re-summarizing it would duplicate the same failing API work (and duplicate warn logs).
+  if (smallMessages.length > 0 && smallMessages.length !== messages.length) {
     try {
       const partialSummary = await summarizeChunks({
         ...params,
