fix: add thinking blocks

strix/agents/base_agent.py

@@ -251,7 +251,7 @@ class BaseAgent(metaclass=AgentMeta):
 
         if self.state.has_waiting_timeout():
             self.state.resume_from_waiting()
-            self.state.add_message("assistant", "Waiting timeout reached. Resuming execution.")
+            self.state.add_message("user", "Waiting timeout reached. Resuming execution.")
 
             from strix.telemetry.tracer import get_global_tracer
 
@@ -364,7 +364,8 @@ class BaseAgent(metaclass=AgentMeta):
             self.state.add_message("user", corrective_message)
             return False
 
-        self.state.add_message("assistant", final_response.content)
+        thinking_blocks = getattr(final_response, "thinking_blocks", None)
+        self.state.add_message("assistant", final_response.content, thinking_blocks=thinking_blocks)
         if tracer:
             tracer.clear_streaming_content(self.state.agent_id)
             tracer.log_chat_message(

strix/agents/state.py

@@ -43,8 +43,11 @@ class AgentState(BaseModel):
         self.iteration += 1
         self.last_updated = datetime.now(UTC).isoformat()
 
-    def add_message(self, role: str, content: Any) -> None:
+    def add_message(self, role: str, content: Any, thinking_blocks: list[dict[str, Any]] | None = None) -> None:
-        self.messages.append({"role": role, "content": content})
+        message = {"role": role, "content": content}
+        if thinking_blocks:
+            message["thinking_blocks"] = thinking_blocks
+        self.messages.append(message)
         self.last_updated = datetime.now(UTC).isoformat()
 
     def add_action(self, action: dict[str, Any]) -> None:

strix/llm/llm.py

@@ -75,6 +75,7 @@ class LLMResponse:
     scan_id: str | None = None
     step_number: int = 1
     role: StepRole = StepRole.AGENT
+    thinking_blocks: list[dict[str, Any]] | None = None  # For reasoning models.
 
 
 @dataclass
@@ -291,12 +292,26 @@ class LLM:
 
         tool_invocations = parse_tool_invocations(accumulated_content)
 
+        # Extract thinking blocks from the complete response if available
+        thinking_blocks = None
+        if chunks and self._should_include_reasoning_effort():
+            complete_response = stream_chunk_builder(chunks)
+            if (
+                hasattr(complete_response, "choices")
+                and complete_response.choices
+                and hasattr(complete_response.choices[0], "message")
+            ):
+                message = complete_response.choices[0].message
+                if hasattr(message, "thinking_blocks") and message.thinking_blocks:
+                    thinking_blocks = message.thinking_blocks
+
         yield LLMResponse(
             scan_id=scan_id,
             step_number=step_number,
             role=StepRole.AGENT,
             content=accumulated_content,
             tool_invocations=tool_invocations if tool_invocations else None,
+            thinking_blocks=thinking_blocks,
         )
 
     def _raise_llm_error(self, e: Exception) -> None: