conductor(checkpoint): Checkpoint end of Phase 1

tests/temp_project.toml

@@ -4,7 +4,6 @@ git_dir = "C:\\projects\\manual_slop"
 system_prompt = ""
 main_context = ""
 word_wrap = true
-summary_only = false
 
 [output]
 output_dir = "./md_gen"
@@ -38,7 +37,8 @@ auto_add = true
 
 [discussion.discussions.main]
 git_commit = ""
-last_updated = "2026-02-23T20:56:57"
+last_updated = "2026-02-23T22:08:38"
 history = [
-    "@2026-02-23T20:12:12\nSystem:\n[PERFORMANCE ALERT] CPU usage high: 121.9%. Please consider optimizing recent changes or reducing load.",
+    "@2026-02-23T22:07:54\nUser:\nHello! This is an automated test. Just say 'Acknowledged'.",
+    "@2026-02-23T22:07:54\nVendor API:\n[GEMINI API ERROR]\n\n404 NOT_FOUND. {'error': {'code': 404, 'message': 'This model models/gemini-2.0-flash is no longer available to new users. Please update your code to use a newer model for the latest features and improvements.', 'status': 'NOT_FOUND'}}",
 ]

tests/test_gui2_mcp.py

@@ -36,35 +36,36 @@ def test_mcp_tool_call_is_dispatched(app_instance):
     # 2. Construct the mock AI response (Gemini format)
     mock_response_with_tool = MagicMock()
     mock_part = MagicMock()
+    mock_part.text = ""
     mock_part.function_call = mock_fc
     mock_candidate = MagicMock()
     mock_candidate.content.parts = [mock_part]
     mock_candidate.finish_reason.name = "TOOL_CALLING"
     mock_response_with_tool.candidates = [mock_candidate]
-    
-    mock_usage_metadata = MagicMock()
-    mock_usage_metadata.prompt_token_count = 100
-    mock_usage_metadata.candidates_token_count = 10
-    mock_usage_metadata.cached_content_token_count = 0
-    mock_response_with_tool.usage_metadata = mock_usage_metadata
+
+    class DummyUsage:
+        prompt_token_count = 100
+        candidates_token_count = 10
+        cached_content_token_count = 0
+
+    mock_response_with_tool.usage_metadata = DummyUsage()
 
     # 3. Create a mock for the final AI response after the tool call
     mock_response_final = MagicMock()
     mock_response_final.text = "Final answer"
     mock_response_final.candidates = []
-    mock_response_final.usage_metadata = mock_usage_metadata
+    mock_response_final.usage_metadata = DummyUsage()
 
     # 4. Patch the necessary components
     with patch("ai_client._ensure_gemini_client"), \
-         patch("ai_client._gemini_client"), \
-         patch("ai_client._gemini_chat") as mock_chat, \
+         patch("ai_client._gemini_client") as mock_client, \
          patch('mcp_client.dispatch', return_value="file content") as mock_dispatch:
-        
+
+        mock_chat = mock_client.chats.create.return_value
         mock_chat.send_message.side_effect = [mock_response_with_tool, mock_response_final]
-        ai_client._gemini_chat = mock_chat
-        
+
         ai_client.set_provider("gemini", "mock-model")
-        
+
         # 5. Call the send function
         ai_client.send(
             md_content="some context",