fix: replace blocking time.sleep with asyncio.sleep in LLM condition

CHANGELOG.md

@@ -6,6 +6,12 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
 
 ## [Unreleased]
 
+### Fixed
+
+- **Blocking sleep in LLM condition** -- Replaced `time.sleep()` with
+  `await asyncio.sleep()` in `LLMAssisted` retry backoff to avoid
+  freezing the event loop during LLM retries.
+
 ### Added
 
 - **Description-diff alignment** -- `DescriptionDiffAlignmentCondition` uses

src/rules/conditions/llm_assisted.py

@@ -5,6 +5,7 @@
 opt-in and clearly documented as having LLM latency in the evaluation path.
 """
 
+import asyncio
 import logging
 import time
 from typing import Any
@@ -218,7 +219,7 @@ async def evaluate(self, context: Any) -> list[Violation]:
                 )
 
                 if attempt < max_attempts:
-                    time.sleep(wait_time)
+                    await asyncio.sleep(wait_time)
                 else:
                     # All attempts failed - gracefully degrade
                     logger.error("All LLM retry attempts exhausted; skipping alignment check.")

tests/unit/rules/conditions/test_llm_assisted.py

@@ -196,7 +196,7 @@ async def test_graceful_degradation_on_llm_failure(self, mock_get_chat_model, co
         assert violations == []
 
     @pytest.mark.asyncio
-    @patch("time.sleep", return_value=None)  # Mock sleep to speed up test
+    @patch("asyncio.sleep", new_callable=AsyncMock)  # Mock async sleep to speed up test
     @patch("src.integrations.providers.get_chat_model")
     async def test_retry_logic_with_exponential_backoff(self, mock_get_chat_model, mock_sleep, condition):
         """When structured invoke fails, retries with exponential backoff."""
@@ -213,7 +213,7 @@ async def test_retry_logic_with_exponential_backoff(self, mock_get_chat_model, m
         # Should have retried 3 times total
         assert mock_structured.ainvoke.await_count == 3
         # Should have slept twice (2s, 4s)
-        assert mock_sleep.call_count == 2
+        assert mock_sleep.await_count == 2
 
     @pytest.mark.asyncio
     @patch("src.integrations.providers.get_chat_model")