fix(llm): add per-turn model request timeout so stalled streams fail fast and retry

strix/config/models.py

@@ -21,6 +21,19 @@ if TYPE_CHECKING:
     from strix.config.settings import Settings
 
 
+def request_timeout_extra_args(timeout_s: float | None) -> dict[str, float] | None:
+    """Per-request model timeout (connect + read/inactivity) as ``extra_args``.
+
+    Restores pre-v1 behavior: a stalled model stream trips this timeout and is
+    retried by ``DEFAULT_MODEL_RETRY`` instead of hanging the agent indefinitely.
+    The value is forwarded to the underlying ``responses.create`` /
+    ``chat.completions.create`` / ``litellm.acompletion`` call.
+    """
+    if not timeout_s or timeout_s <= 0:
+        return None
+    return {"timeout": float(timeout_s)}
+
+
 def _retry_statusless_provider_errors(context: RetryPolicyContext) -> bool:
     """Retry statusless provider errors (e.g. mid-stream quota/billing), but not aborts."""
     normalized = context.normalized

strix/core/inputs.py

@@ -12,6 +12,7 @@ from strix.config.models import (
     DEFAULT_MODEL_RETRY,
     is_known_openai_bare_model,
     model_supports_reasoning,
+    request_timeout_extra_args,
 )
 from strix.core.sessions import scrub_images_from_items
 
@@ -126,11 +127,13 @@ def make_model_settings(
     *,
     model_name: str,
     force_required_tool_choice: bool = False,
+    request_timeout: float | None = None,
 ) -> ModelSettings:
     model_settings = ModelSettings(
         parallel_tool_calls=False,
         retry=DEFAULT_MODEL_RETRY,
         include_usage=True,
+        extra_args=request_timeout_extra_args(request_timeout),
     )
     if (
         reasoning_effort is not None

strix/core/runner.py

@@ -215,6 +215,7 @@ async def run_strix_scan(
             settings.llm.reasoning_effort,
             model_name=resolved_model,
             force_required_tool_choice=settings.llm.force_required_tool_choice,
+            request_timeout=settings.llm.timeout,
         )
         run_config = RunConfig(
             model=resolved_model,

strix/report/dedupe.py

@@ -16,6 +16,7 @@ from strix.config.models import (
     DEFAULT_MODEL_RETRY,
     StrixProvider,
     configure_sdk_model_defaults,
+    request_timeout_extra_args,
 )
 from strix.report.state import get_global_report_state
 
@@ -310,7 +311,11 @@ async def check_duplicate(
         response = await model.get_response(
             system_instructions=DEDUPE_SYSTEM_PROMPT,
             input=user_msg,
-            model_settings=ModelSettings(retry=DEFAULT_MODEL_RETRY, include_usage=True),
+            model_settings=ModelSettings(
+                retry=DEFAULT_MODEL_RETRY,
+                include_usage=True,
+                extra_args=request_timeout_extra_args(settings.llm.timeout),
+            ),
             tools=[],
             output_schema=None,
             handoffs=[],