Rufino Cabrera cd49201a7d feat: Add warning for empty or sentinel LLM responses ... Detect and warn when model returns empty content or sentinel tokens like <|endoftext|> without tool calls or refusals. Changes: - Add _warn_empty_response() helper method (10 lines) - Call helper in get_response() and stream_response() - Add 3 unit tests Non-breaking: warning only, no change to control flow or return types.		2025-12-16 12:07:26 +01:00
..
test_auto_compact.py	Revert "Apply code formatting with make format"	2025-06-14 09:51:19 +00:00
test_openai_chatcompletions.py	feat: Add warning for empty or sentinel LLM responses	2025-12-16 12:07:26 +01:00
test_openai_chatcompletions_converter.py	Revert "Apply code formatting with make format"	2025-06-14 09:51:19 +00:00
test_openai_chatcompletions_stream.py	Fix tests	2025-06-18 10:55:31 +02:00
test_openai_responses_converter.py	fix some test	2025-04-11 08:04:19 +02:00
test_responses.py	fix some test	2025-04-11 08:04:19 +02:00
test_run_config.py	fix function schema test	2025-04-11 08:11:26 +02:00
test_run_step_execution.py	fix function schema test	2025-04-11 08:11:26 +02:00
test_run_step_processing.py	fix function schema test	2025-04-11 08:11:26 +02:00