refactor: lift hardcoded model default + fix stale `is_whitebox` docstring

``"anthropic/claude-sonnet-4-6"`` was duplicated as a kwarg default in
5 places (``run_strix_scan``, ``make_run_config``, ``make_agent_context``,
and twice in ``agents_graph.create_agent``'s ``inner.get(..., default)``
calls). The default was actually dead code: ``validate_environment``
requires ``STRIX_LLM`` to be set before any scan starts, and the CLI/TUI
callers don't pass ``model=`` themselves.

Replaced with a single resolution in ``run_strix_scan``:

    resolved_model = model or load_settings().llm.model
    if not resolved_model:
        raise RuntimeError("No LLM model configured. ...")

then propagated explicitly to ``make_agent_context`` and
``make_run_config``. Both lose their string defaults — ``model`` is now
a required kwarg. The graph tool's ``inner.get("model", "...")`` is
``inner["model"]``: the parent context guarantees it's set.

Drive-by: ``run_strix_scan`` docstring still listed ``is_whitebox`` as
a ``scan_config`` key — stale since ``1e641e5`` derived it from
``targets`` instead. Updated.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

strix/entry.py

@@ -161,7 +161,7 @@ async def run_strix_scan(
     bus: AgentMessageBus | None = None,
     interactive: bool = False,
     max_turns: int = STRIX_DEFAULT_MAX_TURNS,
-    model: str = "anthropic/claude-sonnet-4-6",
+    model: str | None = None,
     cleanup_on_exit: bool = True,
 ) -> RunResult:
     """Run one Strix scan end-to-end against a freshly-prepared sandbox.
@@ -169,7 +169,7 @@ async def run_strix_scan(
     Args:
         scan_config: Per-scan configuration — ``targets``,
             ``user_instructions``, ``diff_scope``, ``scan_mode``,
-            ``is_whitebox``, ``skills``.
+            ``skills``. ``is_whitebox`` is derived from ``targets``.
         scan_id: Used to key the sandbox session cache. Auto-generated
             if omitted — callers that want resume-after-crash semantics
             should pass a stable id.
@@ -181,6 +181,9 @@ async def run_strix_scan(
         interactive: Renders the interactive-mode prompt block on the
             root agent.
         max_turns: Cap on root-agent LLM turns (default 300).
+        model: Litellm model alias. ``None`` (default) reads
+            :attr:`Settings.llm.model` — caller pre-validates via
+            :func:`validate_environment` that it's set.
         cleanup_on_exit: When True (default), tears down the sandbox
             session in a ``finally``. Set to False for resume scenarios
             where the caller wants to preserve the container.
@@ -192,6 +195,12 @@ async def run_strix_scan(
         scan_id = f"scan-{uuid.uuid4().hex[:8]}"
     logger.info("Starting Strix scan %s", scan_id)
 
+    resolved_model = model or load_settings().llm.model
+    if not resolved_model:
+        raise RuntimeError(
+            "No LLM model configured. Set STRIX_LLM env or pass model= to run_strix_scan().",
+        )
+
     # Caller may pre-create the bus so it can hold a handle (e.g., the
     # TUI uses it to route stop / chat-input commands). Otherwise we
     # own the bus internally for the scan's lifetime.
@@ -244,7 +253,7 @@ async def run_strix_scan(
             agent_id=root_id,
             parent_id=None,
             tracer=tracer,
-            model=model,
+            model=resolved_model,
             max_turns=max_turns,
             is_whitebox=is_whitebox,
             diff_scope=diff_scope,
@@ -258,7 +267,7 @@ async def run_strix_scan(
         run_config = make_run_config(
             sandbox_session=bundle["session"],
             sandbox_client=bundle["client"],
-            model=model,
+            model=resolved_model,
             reasoning_effort=reasoning_effort,
         )
 

strix/run_config_factory.py

@@ -55,7 +55,7 @@ DEFAULT_RETRY = ModelRetrySettings(
 def make_run_config(
     *,
     sandbox_session: BaseSandboxSession | None,
-    model: str = "anthropic/claude-sonnet-4-6",
+    model: str,
     reasoning_effort: Literal["low", "medium", "high"] | None = None,
     model_settings_override: ModelSettings | None = None,
     sandbox_client: Any | None = None,
@@ -71,8 +71,8 @@ def make_run_config(
             this scan (one container per scan; see
             :mod:`strix.runtime.session_manager`). ``None`` is allowed
             for unit tests and dry runs.
-        model: Model alias passed to ``MultiProvider``. Defaults to the
-            production Anthropic alias.
+        model: Litellm model alias passed to ``MultiProvider``. Caller
+            resolves from :attr:`Settings.llm.model`.
         reasoning_effort: ``"low" | "medium" | "high"``; routes to
             ``ModelSettings.reasoning``.
         model_settings_override: Optional per-run ``ModelSettings``
@@ -117,7 +117,7 @@ def make_agent_context(
     agent_id: str,
     parent_id: str | None,
     tracer: Any | None,
-    model: str = "anthropic/claude-sonnet-4-6",
+    model: str,
     model_settings: ModelSettings | None = None,
     max_turns: int = 300,
     is_whitebox: bool = False,

strix/tools/agents_graph/tools.py

@@ -446,7 +446,7 @@ async def create_agent(
         agent_id=child_id,
         parent_id=parent_id,
         tracer=inner.get("tracer"),
-        model=inner.get("model", "anthropic/claude-sonnet-4-6"),
+        model=inner["model"],
         model_settings=inner.get("model_settings"),
         max_turns=int(inner.get("max_turns", 300)),
         is_whitebox=bool(inner.get("is_whitebox", False)),
@@ -458,7 +458,7 @@ async def create_agent(
     child_run_config = make_run_config(
         sandbox_session=inner.get("sandbox_session"),
         sandbox_client=inner.get("sandbox_client"),
-        model=inner.get("model", "anthropic/claude-sonnet-4-6"),
+        model=inner["model"],
         model_settings_override=inner.get("model_settings"),
     )