diff --git a/crates/agen/src/engine.rs b/crates/agen/src/engine.rs
index 737fac9a1..45a28293d 100644
--- a/crates/agen/src/engine.rs
+++ b/crates/agen/src/engine.rs
@@ -3401,12 +3401,14 @@ impl<C: LlmClient, S: EngineState, A: Send + Sync> Engine<C, S, A> {
                                 }));
                             }
                         }
-                        None if response_completed => break,
+                        None if !early_dispatch || response_completed => break,
                         None => {
-                            // A clean transport EOF is not an authoritative provider
-                            // response boundary. Treat it exactly like an interrupted
-                            // stream so any early side effect is terminalized and never
-                            // followed by automatic continuation/regeneration.
+                            // Providers that support only response-complete tool calls use
+                            // stream EOF as their established completion boundary. Early
+                            // dispatch is never enabled for them. Once per-block dispatch is
+                            // active, however, transport EOF alone cannot prove that the
+                            // assistant response completed after a side effect may have begun.
+                            // Require the provider's explicit terminal status in that mode.
                             self.timeline.flush_usage();
                             return Ok(StreamResponseOutput {
                                 completion: StreamCompletion::Interrupted {
diff --git a/crates/agen/tests/parallel_execution_test.rs b/crates/agen/tests/parallel_execution_test.rs
index 3638f48d5..279c65eeb 100644
--- a/crates/agen/tests/parallel_execution_test.rs
+++ b/crates/agen/tests/parallel_execution_test.rs
@@ -1679,10 +1679,6 @@ async fn unsupported_provider_visibly_defers_opt_in_dispatch() {
         warning.contains("provider does not expose a trustworthy per-call completion boundary")
     }));
 
-    tx.send(Ok(Event::Status(StatusEvent {
-        status: ResponseStatus::Completed,
-    })))
-    .unwrap();
     drop(tx);
     tokio::time::timeout(Duration::from_secs(1), probe.started.notified())
         .await
diff --git a/crates/agen/tests/response_completion_compatibility_test.rs b/crates/agen/tests/response_completion_compatibility_test.rs
new file mode 100644
index 000000000..c7b7d8fac
--- /dev/null
+++ b/crates/agen/tests/response_completion_compatibility_test.rs
@@ -0,0 +1,142 @@
+use std::sync::{
+    Arc,
+    atomic::{AtomicUsize, Ordering},
+};
+
+use agen::llm_client::{
+    ClientError, LlmClient, Request, ResponseStream, ToolCallCompletionSupport,
+    scheme::{Scheme, gemini::GeminiScheme, openai_chat::OpenAIScheme},
+};
+use agen::{Engine, EngineRunExit, History, ToolCallDispatchMode};
+use async_trait::async_trait;
+
+#[derive(Clone)]
+struct SseFrame {
+    event_type: &'static str,
+    data: &'static str,
+}
+
+#[derive(Clone)]
+struct SchemeReplayClient<S> {
+    scheme: S,
+    responses: Arc<Vec<Vec<SseFrame>>>,
+    stream_calls: Arc<AtomicUsize>,
+}
+
+impl<S> SchemeReplayClient<S> {
+    fn new(scheme: S, responses: Vec<Vec<SseFrame>>) -> Self {
+        Self {
+            scheme,
+            responses: Arc::new(responses),
+            stream_calls: Arc::new(AtomicUsize::new(0)),
+        }
+    }
+
+    fn stream_count(&self) -> usize {
+        self.stream_calls.load(Ordering::SeqCst)
+    }
+}
+
+#[async_trait]
+impl<S: Scheme> LlmClient for SchemeReplayClient<S> {
+    async fn stream(&self, _request: Request) -> Result<ResponseStream, ClientError> {
+        let response_index = self.stream_calls.fetch_add(1, Ordering::SeqCst);
+        let frames = self
+            .responses
+            .get(response_index)
+            .ok_or_else(|| ClientError::Config("scheme replay responses exhausted".to_string()))?;
+        let mut state = S::State::default();
+        let mut output = Vec::new();
+        for frame in frames {
+            match self
+                .scheme
+                .parse_sse(frame.event_type, frame.data, &mut state)
+            {
+                Ok(events) => output.extend(events.into_iter().map(Ok)),
+                Err(error) => {
+                    output.push(Err(error));
+                    break;
+                }
+            }
+        }
+        Ok(Box::pin(futures::stream::iter(output)))
+    }
+
+    fn clone_boxed(&self) -> Box<dyn LlmClient> {
+        Box::new(self.clone())
+    }
+
+    fn tool_call_completion_support(&self) -> ToolCallCompletionSupport {
+        self.scheme.tool_call_completion_support()
+    }
+}
+
+async fn assert_response_complete_scheme_accepts_clean_eof<S: Scheme>(
+    scheme: S,
+    frames: Vec<SseFrame>,
+) {
+    assert_eq!(
+        scheme.tool_call_completion_support(),
+        ToolCallCompletionSupport::ResponseComplete
+    );
+
+    for mode in [
+        ToolCallDispatchMode::AfterResponse,
+        ToolCallDispatchMode::OnToolCallComplete,
+    ] {
+        let client = SchemeReplayClient::new(scheme.clone(), vec![frames.clone()]);
+        let probe = client.clone();
+        let mut engine = Engine::new(client);
+        engine.set_tool_call_dispatch_mode(mode);
+        let mut history = History::new();
+
+        let output = engine.run(&mut history, "hello").await;
+
+        assert!(
+            matches!(output.result, EngineRunExit::Finished),
+            "{mode:?} should accept the response-complete scheme's clean EOF, got {:?}",
+            output.result
+        );
+        assert_eq!(
+            probe.stream_count(),
+            1,
+            "{mode:?} must not open a continuation request after the adapter's normal EOF"
+        );
+    }
+}
+
+#[tokio::test]
+async fn openai_chat_clean_eof_completes_in_default_and_opt_in_fallback_modes() {
+    assert_response_complete_scheme_accepts_clean_eof(
+        OpenAIScheme::new(),
+        vec![
+            SseFrame {
+                event_type: "message",
+                data: r#"{"id":"chatcmpl-123","object":"chat.completion.chunk","created":1694268190,"model":"gpt-4o","choices":[{"index":0,"delta":{"content":"Hello"},"finish_reason":"stop"}]}"#,
+            },
+            SseFrame {
+                event_type: "message",
+                data: "[DONE]",
+            },
+        ],
+    )
+    .await;
+}
+
+#[tokio::test]
+async fn gemini_clean_eof_completes_in_default_and_opt_in_fallback_modes() {
+    assert_response_complete_scheme_accepts_clean_eof(
+        GeminiScheme::new(),
+        vec![
+            SseFrame {
+                event_type: "message",
+                data: r#"{"candidates":[{"content":{"parts":[{"text":"Hello"}],"role":"model"},"finishReason":"STOP","index":0}]}"#,
+            },
+            SseFrame {
+                event_type: "message",
+                data: "[DONE]",
+            },
+        ],
+    )
+    .await;
+}