fix: propagate kimi base-url temperature overrides

Follow up salvaged PR #12668 by threading base_url through the
remaining direct-call sites so kimi-k2.5 uses temperature=1.0 on
api.moonshot.ai and keeps 0.6 on api.kimi.com/coding. Add focused
regression tests for run_agent, trajectory_compressor, and
mini_swe_runner.

tests/run_agent/test_run_agent.py

@@ -918,6 +918,26 @@ class TestBuildApiKwargs:
         assert kwargs["messages"] is messages
         assert kwargs["timeout"] == 1800.0
 
+    def test_public_moonshot_kimi_k2_5_forces_temperature_1(self, agent):
+        agent.base_url = "https://api.moonshot.ai/v1"
+        agent._base_url_lower = agent.base_url.lower()
+        agent.model = "kimi-k2.5"
+        messages = [{"role": "user", "content": "hi"}]
+
+        kwargs = agent._build_api_kwargs(messages)
+
+        assert kwargs["temperature"] == 1.0
+
+    def test_kimi_coding_endpoint_keeps_kimi_k2_5_at_0_6(self, agent):
+        agent.base_url = "https://api.kimi.com/coding/v1"
+        agent._base_url_lower = agent.base_url.lower()
+        agent.model = "kimi-k2.5"
+        messages = [{"role": "user", "content": "hi"}]
+
+        kwargs = agent._build_api_kwargs(messages)
+
+        assert kwargs["temperature"] == 0.6
+
     def test_provider_preferences_injected(self, agent):
         agent.base_url = "https://openrouter.ai/api/v1"
         agent.providers_allowed = ["Anthropic"]

tests/test_mini_swe_runner.py

@@ -26,3 +26,30 @@ def test_run_task_forces_kimi_fixed_temperature():
 
     assert result["completed"] is True
     assert client.chat.completions.create.call_args.kwargs["temperature"] == 0.6
+
+
+def test_run_task_public_moonshot_kimi_k2_5_forces_temperature_1():
+    with patch("openai.OpenAI") as mock_openai:
+        client = MagicMock()
+        client.base_url = "https://api.moonshot.ai/v1"
+        client.chat.completions.create.return_value = SimpleNamespace(
+            choices=[SimpleNamespace(message=SimpleNamespace(content="done", tool_calls=[]))]
+        )
+        mock_openai.return_value = client
+
+        from mini_swe_runner import MiniSWERunner
+
+        runner = MiniSWERunner(
+            model="kimi-k2.5",
+            base_url="https://api.moonshot.ai/v1",
+            api_key="test-key",
+            env_type="local",
+            max_iterations=1,
+        )
+        runner._create_env = MagicMock()
+        runner._cleanup_env = MagicMock()
+
+        result = runner.run_task("2+2")
+
+    assert result["completed"] is True
+    assert client.chat.completions.create.call_args.kwargs["temperature"] == 1.0

tests/test_trajectory_compressor.py

@@ -54,6 +54,30 @@ def test_generate_summary_custom_client_forces_kimi_temperature():
     assert compressor.client.chat.completions.create.call_args.kwargs["temperature"] == 0.6
 
 
+def test_generate_summary_public_moonshot_kimi_k2_5_forces_temperature_1():
+    config = CompressionConfig(
+        summarization_model="kimi-k2.5",
+        base_url="https://api.moonshot.ai/v1",
+        temperature=0.3,
+        summary_target_tokens=100,
+        max_retries=1,
+    )
+    compressor = TrajectoryCompressor.__new__(TrajectoryCompressor)
+    compressor.config = config
+    compressor.logger = MagicMock()
+    compressor._use_call_llm = False
+    compressor.client = MagicMock()
+    compressor.client.chat.completions.create.return_value = SimpleNamespace(
+        choices=[SimpleNamespace(message=SimpleNamespace(content="[CONTEXT SUMMARY]: summary"))]
+    )
+
+    metrics = TrajectoryMetrics()
+    result = compressor._generate_summary("tool output", metrics)
+
+    assert result.startswith("[CONTEXT SUMMARY]:")
+    assert compressor.client.chat.completions.create.call_args.kwargs["temperature"] == 1.0
+
+
 # ---------------------------------------------------------------------------
 # CompressionConfig
 # ---------------------------------------------------------------------------

tests/test_trajectory_compressor_async.py

@@ -141,3 +141,31 @@ async def test_generate_summary_async_custom_client_forces_kimi_temperature():
 
     assert result.startswith("[CONTEXT SUMMARY]:")
     assert async_client.chat.completions.create.call_args.kwargs["temperature"] == 0.6
+
+
+@pytest.mark.asyncio
+async def test_generate_summary_async_public_moonshot_kimi_k2_5_forces_temperature_1():
+    from trajectory_compressor import CompressionConfig, TrajectoryCompressor, TrajectoryMetrics
+
+    config = CompressionConfig(
+        summarization_model="kimi-k2.5",
+        base_url="https://api.moonshot.ai/v1",
+        temperature=0.3,
+        summary_target_tokens=100,
+        max_retries=1,
+    )
+    compressor = TrajectoryCompressor.__new__(TrajectoryCompressor)
+    compressor.config = config
+    compressor.logger = MagicMock()
+    compressor._use_call_llm = False
+    async_client = MagicMock()
+    async_client.chat.completions.create = MagicMock(return_value=SimpleNamespace(
+        choices=[SimpleNamespace(message=SimpleNamespace(content="[CONTEXT SUMMARY]: summary"))]
+    ))
+    compressor._get_async_client = MagicMock(return_value=async_client)
+
+    metrics = TrajectoryMetrics()
+    result = await compressor._generate_summary_async("tool output", metrics)
+
+    assert result.startswith("[CONTEXT SUMMARY]:")
+    assert async_client.chat.completions.create.call_args.kwargs["temperature"] == 1.0