diff --git a/src/adapters/ollama-native.ts b/src/adapters/ollama-native.ts
index 569ae90622..c78f154d49 100644
--- a/src/adapters/ollama-native.ts
+++ b/src/adapters/ollama-native.ts
@@ -306,17 +306,37 @@ function buildNativeMessages(
   // owned by this adapter/request lifecycle rather than process-global state.
   reservedToolCallIds.clear();
   let pending: PendingToolBatch | undefined;
+  // Codex records mid-turn injections (a PostToolUse hook verdict, a context notice) between an
+  // assistant tool call and that call's own tool result. Native Ollama needs the call and its
+  // results adjacent, so those conversational messages wait here instead of closing the batch
+  // early. The openai-chat adapter defers them the same way; refusing the replay killed the turn.
+  let deferred: OllamaNativeMessage[] = [];
+
+  const releaseDeferred = (): void => {
+    if (deferred.length === 0) return;
+    messages.push(...deferred);
+    deferred = [];
+  };
 
   const flushPending = (): void => {
     if (!pending) return;
     for (const call of pending.calls) {
       if (!call.result) {
-        throw new Error(`ollama-native tool call ${call.id} is missing its tool result; refusing interrupted replay`);
+        // No result exists anywhere in the replayed history: the turn was interrupted, or the
+        // result never reached it. State exactly that instead of inventing an outcome, and keep
+        // the conversation replayable.
+        messages.push({
+          role: "tool",
+          tool_call_id: call.id,
+          tool_name: call.wireName,
+          // Same marker text as the chat adapter (openai-chat/messages.ts), so both adapters read
+          // the same in an operator's log. The name is this wire's flattened tool name, which is
+          // what the assistant turn above it carries.
+          content: `[ocx] no tool result was recorded for "${call.wireName}"; execution status unknown — do not treat this as success, failure, or user-provided input.`,
+        });
+        continue;
       }
-    }
-    for (const call of pending.calls) {
-      const result = call.result!;
-      const translated = contentToNative(result.content, "tool result");
+      const translated = contentToNative(call.result.content, "tool result");
       messages.push({
         role: "tool",
         tool_call_id: call.id,
@@ -326,6 +346,7 @@ function buildNativeMessages(
       });
     }
     pending = undefined;
+    releaseDeferred();
   };
 
   for (const message of parsed.context.messages) {
@@ -347,9 +368,22 @@ function buildNativeMessages(
       continue;
     }
 
-    // Native Ollama requires the whole assistant tool-call turn followed by its tool results.  A
-    // new conversational message is a hard boundary; unresolved calls are never fabricated.
-    if (pending) flushPending();
+    // Native Ollama requires the whole assistant tool-call turn followed by its tool results. A
+    // conversational message that arrives while the batch is still open is held aside instead of
+    // closing it, so the call keeps its results adjacent; it is released right after the batch
+    // flushes. Anything else (a new assistant turn) settles the batch first.
+    if (pending) {
+      if (message.role === "user" || message.role === "developer") {
+        const translated = message.role === "user"
+          ? contentToNative(message.content, "user")
+          : contentToNative(message.content, "developer", false);
+        deferred.push(message.role === "user"
+          ? { role: "user", content: translated.content, ...(translated.images ? { images: translated.images } : {}) }
+          : { role: "system", content: translated.content });
+        continue;
+      }
+      flushPending();
+    }
 
     switch (message.role) {
       case "user": {
diff --git a/structure/providers/chat-compat.md b/structure/providers/chat-compat.md
index 9223e4462d..c96f612f43 100644
--- a/structure/providers/chat-compat.md
+++ b/structure/providers/chat-compat.md
@@ -55,6 +55,15 @@ messages until the round completes, reattaching real results to their original c
 and synthesizing explicit "no tool result was recorded" answers only when no real result exists
 (Kimi/Moonshot 400 `ocx-mrqaiw05-269`; unit `devlog/_fin/260718_dangling_toolcall_hardening`).
 
+The native Ollama wire carries the same contract. `src/adapters/ollama-native.ts`
+`buildNativeMessages` defers `user`/`developer` messages that arrive while a batch is open and
+releases them after the tool messages, and answers a call with no result anywhere in the replayed
+history with the same `[ocx] no tool result was recorded for "<name>"` marker. The shape it
+absorbs is ordinary Codex history, not a malformed one: Codex records mid-turn items (a
+`PostToolUse` hook verdict, a context notice) between an assistant `tool_calls` message and that
+call's own result. The strict pair checks (orphan result, duplicate result, result naming another
+tool) still throw on both wires (#4842).
+
 Forward-mode OpenAI passthrough also repairs replayed `call_id` values longer than the Responses
 API's 64-character limit. Sidechat/fork replay can namespace routed-provider ids beyond that limit,
 so each oversized id and all matching call/output items receive the same deterministic,
diff --git a/tests/providers/ollama/ollama-native.test.ts b/tests/providers/ollama/ollama-native.test.ts
index d22cdc1ce4..44a09e8a8c 100644
--- a/tests/providers/ollama/ollama-native.test.ts
+++ b/tests/providers/ollama/ollama-native.test.ts
@@ -261,4 +261,150 @@ describe("ollama-native — request shape", () => {
       { role: "user", content: [{ type: "video", videoUrl: "data:video/mp4;base64,AAAA" }] },
     ]))).toThrow(/cannot send video/);
   });
+
+  test("a mid-turn developer message is deferred instead of closing the tool batch", async () => {
+    const adapter = createOllamaNativeAdapter(ollamaProvider());
+    const { body } = await adapter.buildRequest(parsedWith([
+      { role: "user", content: "continue", timestamp: 0 },
+      {
+        role: "assistant",
+        content: [
+          { type: "text", text: "applying the patch" },
+          { type: "toolCall", id: "call_hook_split", name: "exec", arguments: { cmd: "ls" } },
+        ],
+        timestamp: 1,
+      },
+      { role: "developer", content: "[hook] design findings requiring review", timestamp: 2 },
+      { role: "toolResult", toolCallId: "call_hook_split", toolName: "exec", content: "done", isError: false, timestamp: 3 },
+    ]));
+    const messages = JSON.parse(String(body)).messages;
+    expect(messages.map((message: { role: string }) => message.role))
+      .toEqual(["user", "assistant", "tool", "system"]);
+    expect(messages[2].tool_call_id).toBe("call_hook_split");
+    expect(messages[2].content).toBe("done");
+    expect(messages[3].content).toBe("[hook] design findings requiring review");
+  });
+
+  test("a deferred user message keeps its text and images after the tool result", async () => {
+    const adapter = createOllamaNativeAdapter(ollamaProvider());
+    const png = "data:image/png;base64,iVBORw0KGgo=";
+    const { body } = await adapter.buildRequest(parsedWith([
+      { role: "user", content: "start", timestamp: 0 },
+      {
+        role: "assistant",
+        content: [{ type: "toolCall", id: "call_mid_user", name: "exec", arguments: { cmd: "ls" } }],
+        timestamp: 1,
+      },
+      {
+        role: "user",
+        content: [{ type: "text", text: "look at this" }, { type: "image", imageUrl: png }],
+        timestamp: 2,
+      },
+      { role: "toolResult", toolCallId: "call_mid_user", toolName: "exec", content: "done", isError: false, timestamp: 3 },
+    ]));
+    const messages = JSON.parse(String(body)).messages;
+    expect(messages.map((message: { role: string }) => message.role))
+      .toEqual(["user", "assistant", "tool", "user"]);
+    expect(messages[2].tool_call_id).toBe("call_mid_user");
+    expect(messages[2].content).toBe("done");
+    expect(messages[3].content).toBe("look at this");
+    expect(messages[3].images).toEqual(["iVBORw0KGgo="]);
+  });
+
+  test("out-of-order results inside a parallel batch still serialize in call order", async () => {
+    const adapter = createOllamaNativeAdapter(ollamaProvider());
+    const { body } = await adapter.buildRequest(parsedWith([
+      { role: "user", content: "continue", timestamp: 0 },
+      {
+        role: "assistant",
+        content: [
+          { type: "toolCall", id: "call_first", name: "exec", arguments: { cmd: "ls" } },
+          { type: "toolCall", id: "call_second", name: "exec", arguments: { cmd: "pwd" } },
+        ],
+        timestamp: 1,
+      },
+      { role: "developer", content: "[hook] findings", timestamp: 2 },
+      { role: "toolResult", toolCallId: "call_second", toolName: "exec", content: "second", isError: false, timestamp: 3 },
+      { role: "toolResult", toolCallId: "call_first", toolName: "exec", content: "first", isError: false, timestamp: 4 },
+    ]));
+    const messages = JSON.parse(String(body)).messages;
+    expect(messages.map((message: { role: string }) => message.role))
+      .toEqual(["user", "assistant", "tool", "tool", "system"]);
+    expect(messages[2].tool_call_id).toBe("call_first");
+    expect(messages[2].content).toBe("first");
+    expect(messages[3].tool_call_id).toBe("call_second");
+    expect(messages[3].content).toBe("second");
+  });
+
+  test("a call with no recorded result answers with an explicit unknown status", async () => {
+    const adapter = createOllamaNativeAdapter(ollamaProvider());
+    const { body } = await adapter.buildRequest(parsedWith([
+      { role: "user", content: "start", timestamp: 0 },
+      {
+        role: "assistant",
+        content: [{ type: "toolCall", id: "call_interrupted", name: "exec", arguments: { cmd: "ls" } }],
+        timestamp: 1,
+      },
+      { role: "user", content: "continue", timestamp: 2 },
+    ]));
+    const messages = JSON.parse(String(body)).messages;
+    expect(messages.map((message: { role: string }) => message.role))
+      .toEqual(["user", "assistant", "tool", "user"]);
+    expect(messages[2].tool_call_id).toBe("call_interrupted");
+    expect(messages[2].content).toContain("no tool result was recorded");
+    expect(messages[2].content).toContain('"exec"');
+    expect(messages[2].content).toContain("do not treat this as success, failure, or user-provided input");
+    expect(messages[3].content).toBe("continue");
+  });
+
+  test("a second assistant turn settles the first batch before its own", async () => {
+    const adapter = createOllamaNativeAdapter(ollamaProvider());
+    const { body } = await adapter.buildRequest(parsedWith([
+      { role: "user", content: "start", timestamp: 0 },
+      {
+        role: "assistant",
+        content: [{ type: "toolCall", id: "call_first", name: "exec", arguments: { cmd: "ls" } }],
+        timestamp: 1,
+      },
+      { role: "developer", content: "[hook] findings", timestamp: 2 },
+      { role: "toolResult", toolCallId: "call_first", toolName: "exec", content: "first done", isError: false, timestamp: 3 },
+      {
+        role: "assistant",
+        content: [{ type: "toolCall", id: "call_second", name: "exec", arguments: { cmd: "pwd" } }],
+        timestamp: 4,
+      },
+      { role: "toolResult", toolCallId: "call_second", toolName: "exec", content: "second done", isError: false, timestamp: 5 },
+    ]));
+    const messages = JSON.parse(String(body)).messages;
+    expect(messages.map((message: { role: string }) => message.role))
+      .toEqual(["user", "assistant", "tool", "system", "assistant", "tool"]);
+    expect(messages[2].tool_call_id).toBe("call_first");
+    expect(messages[2].content).toBe("first done");
+    expect(messages[3].content).toBe("[hook] findings");
+    expect(messages[4].tool_calls[0].id).toBe("call_second");
+    expect(messages[5].tool_call_id).toBe("call_second");
+    expect(messages[5].content).toBe("second done");
+  });
+
+  test("an orphan tool result is still refused", () => {
+    const adapter = createOllamaNativeAdapter(ollamaProvider());
+    expect(() => adapter.buildRequest(parsedWith([
+      { role: "user", content: "hi" },
+      { role: "toolResult", toolCallId: "call_ghost", toolName: "exec", content: "x", isError: false, timestamp: 1 },
+    ]))).toThrow(/orphan tool result/);
+  });
+
+  test("a duplicate result for the same call is still refused", () => {
+    const adapter = createOllamaNativeAdapter(ollamaProvider());
+    expect(() => adapter.buildRequest(parsedWith([
+      { role: "user", content: "hi" },
+      {
+        role: "assistant",
+        content: [{ type: "toolCall", id: "call_once", name: "exec", arguments: { cmd: "ls" } }],
+        timestamp: 1,
+      },
+      { role: "toolResult", toolCallId: "call_once", toolName: "exec", content: "once", isError: false, timestamp: 2 },
+      { role: "toolResult", toolCallId: "call_once", toolName: "exec", content: "twice", isError: false, timestamp: 3 },
+    ]))).toThrow(/duplicate tool result/);
+  });
 });