@vellumai/assistant 0.8.5 → 0.8.7

package/src/__tests__/openai-responses-provider.test.ts

@@ -47,7 +47,7 @@ mock.module("openai", () => ({
       lastConstructorOptions = opts;
     }
     responses = {
-      stream: (
+      create: async (
         params: Record<string, unknown>,
         options?: Record<string, unknown>,
       ) => {
@@ -165,6 +165,26 @@ function completedEvent(
   };
 }
 
+function incompleteEvent(
+  reason: "max_output_tokens" | "content_filter",
+  inputTokens: number,
+  outputTokens: number,
+): FakeStreamEvent {
+  return {
+    type: "response.incomplete",
+    response: {
+      model: "gpt-5.2",
+      status: "incomplete",
+      incomplete_details: { reason },
+      output: [],
+      usage: {
+        input_tokens: inputTokens,
+        output_tokens: outputTokens,
+      },
+    },
+  };
+}
+
 // ---------------------------------------------------------------------------
 // Tests
 // ---------------------------------------------------------------------------
@@ -243,6 +263,21 @@ describe("OpenAIResponsesProvider", () => {
     expect(result.stopReason).toBe("stop");
   });
 
+  test("maps response.incomplete max output details to stopReason", async () => {
+    fakeStreamEvents = [
+      textDeltaEvent("Partial"),
+      incompleteEvent("max_output_tokens", 12, 8),
+    ];
+
+    const result = await provider.sendMessage([
+      { role: "user", content: [{ type: "text", text: "Hi" }] },
+    ]);
+
+    expect(result.content).toEqual([{ type: "text", text: "Partial" }]);
+    expect(result.usage).toEqual({ inputTokens: 12, outputTokens: 8 });
+    expect(result.stopReason).toBe("max_output_tokens");
+  });
+
   // -----------------------------------------------------------------------
   // Streaming events
   // -----------------------------------------------------------------------
@@ -256,8 +291,6 @@ describe("OpenAIResponsesProvider", () => {
     const events: ProviderEvent[] = [];
     await provider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      undefined,
       { onEvent: (e) => events.push(e) },
     );
 
@@ -274,8 +307,7 @@ describe("OpenAIResponsesProvider", () => {
 
     await provider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      "You are a helpful assistant.",
+      { systemPrompt: "You are a helpful assistant." },
     );
 
     expect(lastStreamParams!.instructions).toBe("You are a helpful assistant.");
@@ -307,7 +339,7 @@ describe("OpenAIResponsesProvider", () => {
 
     await provider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Read /tmp/test" }] }],
-      tools,
+      { tools: tools },
     );
 
     const sentTools = lastStreamParams!.tools as Array<Record<string, unknown>>;
@@ -512,8 +544,6 @@ describe("OpenAIResponsesProvider", () => {
 
     await provider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      undefined,
       { config: { max_tokens: 64000 } },
     );
 
@@ -528,8 +558,6 @@ describe("OpenAIResponsesProvider", () => {
 
     await provider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      undefined,
       { config: { effort: "high" } },
     );
 
@@ -541,8 +569,6 @@ describe("OpenAIResponsesProvider", () => {
 
     await provider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      undefined,
       { config: { effort: "max" } },
     );
 
@@ -554,8 +580,6 @@ describe("OpenAIResponsesProvider", () => {
 
     await provider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      undefined,
       { config: { effort: "xhigh" } },
     );
 
@@ -567,8 +591,6 @@ describe("OpenAIResponsesProvider", () => {
 
     await provider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      undefined,
       { config: {} },
     );
 
@@ -583,8 +605,6 @@ describe("OpenAIResponsesProvider", () => {
 
     await provider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      undefined,
       { config: { effort: "none" } },
     );
 
@@ -599,8 +619,6 @@ describe("OpenAIResponsesProvider", () => {
 
     await provider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      undefined,
       { config: { verbosity: "low" } },
     );
 
@@ -612,8 +630,6 @@ describe("OpenAIResponsesProvider", () => {
 
     await provider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      undefined,
       { config: { verbosity: "high" } },
     );
 
@@ -625,8 +641,6 @@ describe("OpenAIResponsesProvider", () => {
 
     await provider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      undefined,
       { config: {} },
     );
 
@@ -637,10 +651,7 @@ describe("OpenAIResponsesProvider", () => {
     fakeStreamEvents = [textDeltaEvent("OK"), completedEvent(10, 2)];
 
     await provider.sendMessage(
-      [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      undefined,
-      // Cast through unknown — the test deliberately exercises the runtime
+      [{ role: "user", content: [{ type: "text", text: "Hi" }] }], // Cast through unknown — the test deliberately exercises the runtime
       // guard against malformed values that would bypass the type system
       // (e.g. arriving via the index signature on `SendMessageConfig`).
       { config: { verbosity: "bogus" as unknown as "low" } },
@@ -657,8 +668,6 @@ describe("OpenAIResponsesProvider", () => {
 
     await oSeriesProvider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      undefined,
       { config: { verbosity: "high" } },
     );
 
@@ -671,8 +680,6 @@ describe("OpenAIResponsesProvider", () => {
 
     await oSeriesProvider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      undefined,
       { config: { verbosity: "high", model: "gpt-5.2" } },
     );
 
@@ -688,8 +695,6 @@ describe("OpenAIResponsesProvider", () => {
 
     await ftProvider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      undefined,
       { config: { verbosity: "low" } },
     );
 
@@ -707,8 +712,6 @@ describe("OpenAIResponsesProvider", () => {
 
     await provider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      undefined,
       { config: { model: "gpt-5.4" } },
     );
 
@@ -951,8 +954,6 @@ describe("OpenAIResponsesProvider", () => {
 
     await provider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      undefined,
       { signal: controller.signal },
     );
 
@@ -968,8 +969,6 @@ describe("OpenAIResponsesProvider", () => {
 
     await provider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      undefined,
       { signal: controller.signal },
     );
 
@@ -1041,8 +1040,6 @@ describe("OpenAIResponsesProvider", () => {
     try {
       await provider.sendMessage(
         [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-        undefined,
-        undefined,
         { signal: controller.signal },
       );
       expect(true).toBe(false);
@@ -1064,8 +1061,6 @@ describe("OpenAIResponsesProvider", () => {
     try {
       await provider.sendMessage(
         [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-        undefined,
-        undefined,
         { signal: controller.signal },
       );
       expect(true).toBe(false);
@@ -1083,8 +1078,6 @@ describe("OpenAIResponsesProvider", () => {
     try {
       await provider.sendMessage(
         [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-        undefined,
-        undefined,
         { signal: controller.signal },
       );
       expect(true).toBe(false);
@@ -1116,11 +1109,10 @@ describe("OpenAIResponsesProvider", () => {
 
     const result = await provider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      "System prompt",
+      { systemPrompt: "System prompt" },
     );
 
-    // rawRequest should contain the params sent to responses.stream()
+    // rawRequest should contain the params sent to responses.create()
     const rawReq = result.rawRequest as Record<string, unknown>;
     expect(rawReq.model).toBe("gpt-5.2");
     expect(rawReq.instructions).toBe("System prompt");
@@ -1328,7 +1320,7 @@ describe("OpenAIResponsesProvider — Native Web Search", () => {
 
     await nativeProvider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Search for cats" }] }],
-      [webSearchTool],
+      { tools: [webSearchTool] },
     );
 
     const sentTools = lastStreamParams!.tools as Array<Record<string, unknown>>;
@@ -1342,7 +1334,7 @@ describe("OpenAIResponsesProvider — Native Web Search", () => {
 
     await nonNativeProvider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Search for cats" }] }],
-      [webSearchTool],
+      { tools: [webSearchTool] },
     );
 
     const sentTools = lastStreamParams!.tools as Array<Record<string, unknown>>;
@@ -1367,7 +1359,7 @@ describe("OpenAIResponsesProvider — Native Web Search", () => {
 
     await nativeProvider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Search and read" }] }],
-      [fileReadTool, webSearchTool],
+      { tools: [fileReadTool, webSearchTool] },
     );
 
     const sentTools = lastStreamParams!.tools as Array<Record<string, unknown>>;
@@ -1396,7 +1388,7 @@ describe("OpenAIResponsesProvider — Native Web Search", () => {
 
     await nativeProvider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Read file" }] }],
-      [fileReadTool],
+      { tools: [fileReadTool] },
     );
 
     const sentTools = lastStreamParams!.tools as Array<Record<string, unknown>>;
@@ -1431,9 +1423,7 @@ describe("OpenAIResponsesProvider — Native Web Search", () => {
     const events: ProviderEvent[] = [];
     await nativeProvider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Search for cats" }] }],
-      [webSearchTool],
-      undefined,
-      { onEvent: (e) => events.push(e) },
+      { tools: [webSearchTool], onEvent: (e) => events.push(e) },
     );
 
     const startEvents = events.filter((e) => e.type === "server_tool_start");
@@ -1459,9 +1449,7 @@ describe("OpenAIResponsesProvider — Native Web Search", () => {
     const events: ProviderEvent[] = [];
     await nativeProvider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Search for dogs" }] }],
-      [webSearchTool],
-      undefined,
-      { onEvent: (e) => events.push(e) },
+      { tools: [webSearchTool], onEvent: (e) => events.push(e) },
     );
 
     const completeEvents = events.filter(
@@ -1489,9 +1477,7 @@ describe("OpenAIResponsesProvider — Native Web Search", () => {
     const events: ProviderEvent[] = [];
     await nativeProvider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Search multiple" }] }],
-      [webSearchTool],
-      undefined,
-      { onEvent: (e) => events.push(e) },
+      { tools: [webSearchTool], onEvent: (e) => events.push(e) },
     );
 
     const startEvents = events.filter((e) => e.type === "server_tool_start");
@@ -1539,9 +1525,7 @@ describe("OpenAIResponsesProvider — Native Web Search", () => {
     const events: ProviderEvent[] = [];
     await nativeProvider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Read file" }] }],
-      [fileReadTool],
-      undefined,
-      { onEvent: (e) => events.push(e) },
+      { tools: [fileReadTool], onEvent: (e) => events.push(e) },
     );
 
     const serverToolEvents = events.filter(
@@ -1567,7 +1551,7 @@ describe("OpenAIResponsesProvider — Native Web Search", () => {
 
     const result = await nativeProvider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Search for cats" }] }],
-      [webSearchTool],
+      { tools: [webSearchTool] },
     );
 
     // server_tool_use + web_search_tool_result pair should appear before text
@@ -1607,7 +1591,7 @@ describe("OpenAIResponsesProvider — Native Web Search", () => {
           content: [{ type: "text", text: "Search two things" }],
         },
       ],
-      [webSearchTool],
+      { tools: [webSearchTool] },
     );
 
     expect(result.content).toHaveLength(5);
@@ -1650,7 +1634,7 @@ describe("OpenAIResponsesProvider — Native Web Search", () => {
 
     const result = await nativeProvider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hello" }] }],
-      [webSearchTool],
+      { tools: [webSearchTool] },
     );
 
     expect(result.content).toHaveLength(1);
@@ -1671,15 +1655,13 @@ describe("OpenAIResponsesProvider — Native Web Search", () => {
 
     await codexProvider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      undefined,
       { config: { max_tokens: 64000 } },
     );
 
     expect(lastStreamParams!.max_output_tokens).toBeUndefined();
   });
 
-  test("codexSubscription: strips reasoning param even when effort is set", async () => {
+  test("codexSubscription: forwards reasoning param when effort is set", async () => {
     const codexProvider = new OpenAIResponsesProvider("sk-test", "gpt-5.4", {
       codexSubscription: true,
     });
@@ -1687,15 +1669,13 @@ describe("OpenAIResponsesProvider — Native Web Search", () => {
 
     await codexProvider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      undefined,
       { config: { effort: "high" } },
     );
 
-    expect(lastStreamParams!.reasoning).toBeUndefined();
+    expect(lastStreamParams!.reasoning).toEqual({ effort: "high" });
   });
 
-  test("codexSubscription: strips text.verbosity param", async () => {
+  test("codexSubscription: forwards text.verbosity param", async () => {
     const codexProvider = new OpenAIResponsesProvider("sk-test", "gpt-5.4", {
       codexSubscription: true,
     });
@@ -1703,12 +1683,10 @@ describe("OpenAIResponsesProvider — Native Web Search", () => {
 
     await codexProvider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      undefined,
       { config: { verbosity: "low" } },
     );
 
-    expect(lastStreamParams!.text).toBeUndefined();
+    expect(lastStreamParams!.text).toEqual({ verbosity: "low" });
   });
 
   test("codexSubscription: uses Codex baseURL", async () => {
@@ -1721,7 +1699,7 @@ describe("OpenAIResponsesProvider — Native Web Search", () => {
     );
   });
 
-  test("codexSubscription: strips tools param", async () => {
+  test("codexSubscription: forwards tools param", async () => {
     const codexProvider = new OpenAIResponsesProvider("sk-test", "gpt-5.4", {
       codexSubscription: true,
     });
@@ -1734,10 +1712,35 @@ describe("OpenAIResponsesProvider — Native Web Search", () => {
     };
     await codexProvider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      [sampleTool],
+      { tools: [sampleTool] },
+    );
+
+    expect(lastStreamParams!.tools).toEqual([
+      {
+        type: "function",
+        name: "test_tool",
+        description: "A test tool",
+        parameters: { type: "object", properties: {} },
+        strict: null,
+      },
+    ]);
+  });
+
+  test("codexSubscription: maps web_search to the Codex native web_search tool", async () => {
+    const codexProvider = new OpenAIResponsesProvider("sk-test", "gpt-5.4", {
+      codexSubscription: true,
+      useNativeWebSearch: true,
+    });
+    fakeStreamEvents = [textDeltaEvent("OK"), completedEvent(10, 2)];
+
+    await codexProvider.sendMessage(
+      [{ role: "user", content: [{ type: "text", text: "Search for cats" }] }],
+      { tools: [webSearchTool] },
     );
 
-    expect(lastStreamParams!.tools).toBeUndefined();
+    expect(lastStreamParams!.tools).toEqual([
+      { type: "web_search", external_web_access: false },
+    ]);
   });
 
   test("codexSubscription: still sends model, input, and instructions", async () => {
@@ -1748,15 +1751,16 @@ describe("OpenAIResponsesProvider — Native Web Search", () => {
 
     await codexProvider.sendMessage(
       [{ role: "user", content: [{ type: "text", text: "Hi" }] }],
-      undefined,
-      "You are helpful.",
-      { config: { effort: "max", verbosity: "high", max_tokens: 64000 } },
+      {
+        systemPrompt: "You are helpful.",
+        config: { effort: "max", verbosity: "high", max_tokens: 64000 },
+      },
     );
 
     expect(lastStreamParams!.model).toBe("gpt-5.4");
     expect(lastStreamParams!.instructions).toBe("You are helpful.");
     expect(lastStreamParams!.max_output_tokens).toBeUndefined();
-    expect(lastStreamParams!.reasoning).toBeUndefined();
-    expect(lastStreamParams!.text).toBeUndefined();
+    expect(lastStreamParams!.reasoning).toEqual({ effort: "xhigh" });
+    expect(lastStreamParams!.text).toEqual({ verbosity: "high" });
   });
 });

package/src/__tests__/openrouter-provider-only.test.ts

@@ -101,19 +101,19 @@ describe("OpenRouter provider.only plumbing", () => {
         config: { openrouter: { only: ["xAI"] } },
       });
       expect(extras).toEqual({
-        reasoning: { enabled: false },
         provider: { only: ["xAI"] },
       });
     });
 
-    test("omits provider when openrouter.only is absent", () => {
+    test("omits reasoning and provider when config is empty", () => {
       const provider = new ProbeOpenRouterProvider(
         "fake-key",
         "x-ai/grok-4.20-beta",
       );
       const extras = provider.probeExtras({ config: {} });
-      expect(extras).toEqual({ reasoning: { enabled: false } });
+      expect(extras).toEqual({});
       expect(extras.provider).toBe(undefined);
+      expect(extras.reasoning).toBe(undefined);
     });
 
     test("enables thinking with default detailed summary alongside provider.only", () => {
@@ -133,7 +133,7 @@ describe("OpenRouter provider.only plumbing", () => {
       });
     });
 
-    test("disabled thinking keeps reasoning disabled and omits summary", () => {
+    test("disabled thinking omits reasoning entirely", () => {
       const provider = new ProbeOpenRouterProvider(
         "fake-key",
         "x-ai/grok-4.20-beta",
@@ -145,9 +145,9 @@ describe("OpenRouter provider.only plumbing", () => {
         },
       });
       expect(extras).toEqual({
-        reasoning: { enabled: false },
         provider: { only: ["xAI"] },
       });
+      expect(extras.reasoning).toBe(undefined);
     });
 
     test("nests effort under reasoning and maps `max` to xhigh", () => {
@@ -182,6 +182,28 @@ describe("OpenRouter provider.only plumbing", () => {
       });
     });
 
+    test("effort without thinking does not emit reasoning", () => {
+      const provider = new ProbeOpenRouterProvider(
+        "fake-key",
+        "x-ai/grok-4.20-beta",
+      );
+      const extras = provider.probeExtras({
+        config: { thinking: { type: "disabled" }, effort: "low" },
+      });
+      expect(extras.reasoning).toBe(undefined);
+    });
+
+    test("omitting reasoning avoids 400 from reasoning-only models like DeepSeek R1", () => {
+      const provider = new ProbeOpenRouterProvider(
+        "fake-key",
+        "deepseek/deepseek-r1-0528",
+      );
+      const extras = provider.probeExtras({
+        config: { thinking: { type: "disabled" } },
+      });
+      expect(extras.reasoning).toBe(undefined);
+    });
+
     test("ignores an invalid summary override and falls back to detailed", () => {
       const provider = new ProbeOpenRouterProvider(
         "fake-key",

package/src/__tests__/outbound-slack-persistence.test.ts

@@ -76,8 +76,9 @@ mock.module("../memory/conversation-crud.js", () => ({
     conversationId: string,
     role: string,
     content: string,
-    metadata?: Record<string, unknown>,
+    options?: { metadata?: Record<string, unknown> },
   ) => {
+    const metadata = options?.metadata;
     const id = `mock-msg-${addMessageCalls.length + 1}`;
     addMessageCalls.push({ id, conversationId, role, content, metadata });
     persistedRows.push({
@@ -107,10 +108,47 @@ mock.module("../memory/conversation-crud.js", () => ({
         : {};
     row.metadata = JSON.stringify({ ...existing, ...updates });
   },
-  updateMessageContent: () => {},
+  updateMessageContent: (messageId: string, content: string) => {
+    // Mirror updateContent into the same capture array so existing
+    // `lastAssistantPersisted()` assertions continue to find the row that
+    // was reserved at `llm_call_started` time.
+    const row = persistedRows.find((candidate) => candidate.id === messageId);
+    if (row) row.content = content;
+    const call = addMessageCalls.find((c) => c.id === messageId);
+    if (call) call.content = content;
+  },
   // The handler treats provenance as a flat spread; returning {} keeps the
   // metadata snapshot focused on the fields under test.
   provenanceFromTrustContext: () => ({}),
+  reserveMessage: mock(
+    async (
+      conversationId: string,
+      role: string,
+      metadata?: Record<string, unknown>,
+    ) => {
+      // B3: production code creates the assistant row at `llm_call_started`
+      // via `reserveMessage`, stamping channel metadata at reserve time.
+      // Mirror that into the addMessage capture array so existing
+      // `lastAssistantPersisted()` assertions keep working.
+      const id = `mock-msg-${addMessageCalls.length + 1}-reserve`;
+      addMessageCalls.push({
+        id,
+        conversationId,
+        role,
+        content: "",
+        metadata,
+      });
+      persistedRows.push({
+        id,
+        conversationId,
+        role,
+        content: "",
+        createdAt: Date.now(),
+        metadata: metadata ? JSON.stringify(metadata) : null,
+      });
+      return { id };
+    },
+  ),
 }));
 
 mock.module("../memory/llm-request-log-store.js", () => ({
@@ -147,6 +185,7 @@ import type {
 } from "../daemon/conversation-agent-loop-handlers.js";
 import {
   createEventHandlerState,
+  handleLlmCallStarted,
   handleMessageComplete,
 } from "../daemon/conversation-agent-loop-handlers.js";
 import type { ServerMessage } from "../daemon/message-protocol.js";
@@ -255,6 +294,7 @@ describe("outbound assistant Slack metadata persistence", () => {
       assistantMessageChannel: "slack",
       requesterChatId: channelId,
     });
+    await handleLlmCallStarted(state, deps);
     await handleMessageComplete(state, deps, makeMessageCompleteEvent("hi"));
 
     const persisted = lastAssistantPersisted();
@@ -296,6 +336,7 @@ describe("outbound assistant Slack metadata persistence", () => {
       requesterTimezoneLabel: "ET",
       clientTimezone: "America/Los_Angeles",
     });
+    await handleLlmCallStarted(state, deps);
     await handleMessageComplete(
       state,
       deps,
@@ -324,6 +365,7 @@ describe("outbound assistant Slack metadata persistence", () => {
       requesterChatId: channelId,
       clientTimezone: "America/Los_Angeles",
     });
+    await handleLlmCallStarted(state, deps);
     await handleMessageComplete(
       state,
       deps,
@@ -351,6 +393,7 @@ describe("outbound assistant Slack metadata persistence", () => {
       requesterChatId: channelId,
       requesterTimezoneLabel: "ET",
     });
+    await handleLlmCallStarted(state, deps);
     await handleMessageComplete(
       state,
       deps,
@@ -394,6 +437,7 @@ describe("outbound assistant Slack metadata persistence", () => {
       assistantMessageChannel: "slack",
       requesterChatId: channelId,
     });
+    await handleLlmCallStarted(state, deps);
     await handleMessageComplete(state, deps, makeMessageCompleteEvent("hello"));
 
     const persisted = lastAssistantPersisted();
@@ -425,6 +469,7 @@ describe("outbound assistant Slack metadata persistence", () => {
       assistantMessageChannel: "slack",
       requesterChatId: channelId,
     });
+    await handleLlmCallStarted(state, deps);
     await handleMessageComplete(
       state,
       deps,
@@ -446,6 +491,7 @@ describe("outbound assistant Slack metadata persistence", () => {
     const deps = makeDeps(conversationId, {
       assistantMessageChannel: "vellum",
     });
+    await handleLlmCallStarted(state, deps);
     await handleMessageComplete(
       state,
       deps,

package/src/__tests__/overflow-reduce-pipeline.test.ts

@@ -33,10 +33,8 @@ import {
 } from "../daemon/context-overflow-reducer.js";
 import type { InjectionMode } from "../daemon/conversation-runtime-assembly.js";
 import type { TrustContext } from "../daemon/trust-context.js";
-import {
-  defaultOverflowReduceMiddleware,
-  defaultOverflowReducePlugin,
-} from "../plugins/defaults/overflow-reduce.js";
+import defaultOverflowReduceMiddleware from "../plugins/defaults/overflow-reduce/middlewares/overflowReduce.js";
+import { defaultOverflowReducePlugin } from "../plugins/defaults/overflow-reduce/register.js";
 import { runPipeline } from "../plugins/pipeline.js";
 import {
   getMiddlewaresFor,