feat(core): opt-in parallel tool calls via STRIX_PARALLEL_TOOL_CALLS

strix/agents/factory.py

@@ -717,6 +717,7 @@ def build_strix_agent(
     is_whitebox: bool = False,
     is_diff_scoped: bool = False,
     interactive: bool = False,
+    parallel_tool_calls: bool = False,
     chat_completions_tools: bool = False,
     strict_tool_schemas: bool = True,
     system_prompt_context: dict[str, Any] | None = None,
@@ -751,6 +752,7 @@ def build_strix_agent(
             is_root=is_root,
             is_diff_scoped=is_diff_scoped,
             interactive=interactive,
+            parallel_tool_calls=parallel_tool_calls,
             system_prompt_context=system_prompt_context,
             supports_images=supports_images,
             report_tool=report_tool.name,
@@ -813,6 +815,7 @@ def make_child_factory(
     is_whitebox: bool = False,
     is_diff_scoped: bool = False,
     interactive: bool = False,
+    parallel_tool_calls: bool = False,
     chat_completions_tools: bool = False,
     strict_tool_schemas: bool = True,
     system_prompt_context: dict[str, Any] | None = None,
@@ -835,6 +838,7 @@ def make_child_factory(
             is_whitebox=is_whitebox,
             is_diff_scoped=is_diff_scoped,
             interactive=interactive,
+            parallel_tool_calls=parallel_tool_calls,
             chat_completions_tools=chat_completions_tools,
             strict_tool_schemas=strict_tool_schemas,
             system_prompt_context=system_prompt_context,

strix/agents/prompt.py

@@ -83,6 +83,7 @@ def render_system_prompt(
     is_root: bool = False,
     is_diff_scoped: bool = False,
     interactive: bool = False,
+    parallel_tool_calls: bool = False,
     system_prompt_context: dict[str, Any] | None = None,
     include_scope: bool = True,
     supports_images: bool = True,
@@ -131,6 +132,7 @@ def render_system_prompt(
             requested_skill_names=[name for name in skill_content if name not in shared],
             available_skills=get_available_skills(),
             interactive=interactive,
+            parallel_tool_calls=parallel_tool_calls,
             is_root=is_root,
             system_prompt_context=system_prompt_context or {},
             include_scope=include_scope,

strix/agents/prompts/system_prompt.jinja

@@ -44,7 +44,7 @@ INTERACTIVE BEHAVIOR:
 - If you end a turn on plain text and the nudge arrives, your words already reached the user. Do not repeat them: call wait_for_user.
 - You can narrate while you work — plain text is shown to the user as you go. Narrating is free; wait_for_user is specifically the act of WAITING for the user, so do not call it just to give a status update.
 - Respond naturally when the user asks questions or gives instructions.
-- While actively working on a task, every turn should carry exactly one tool call — use think to plan, the appropriate tool to act, and wait_for_user only when you genuinely need the user.
+- While actively working on a task, every turn should carry {% if parallel_tool_calls %}a tool call (independent calls can go in the same turn){% else %}exactly one tool call{% endif %} — use think to plan, the appropriate tool to act, and wait_for_user only when you genuinely need the user.
 - Never loop through think or other tools just to prepare, polish, confirm, or announce an answer. Once you know the answer, write it and call wait_for_user.
 {% else %}
 AUTONOMOUS BEHAVIOR:
@@ -55,6 +55,9 @@ AUTONOMOUS BEHAVIOR:
 - There is no user attached to this run, so there is nobody to ask and nothing to yield to. If there is nothing left to execute: do NOT send filler/repetitive text — either call wait_for_agents (only if you are genuinely expecting another agent to message you) or finish your work (subagents: agent_finish; root: finish_scan)
 - While the agent loop is running, almost every output MUST be a tool call. Do NOT send plain text messages; act via tools. If waiting on another agent, use wait_for_agents; when done, use agent_finish (subagents) or finish_scan (root)
 - A text-only turn does nothing: it neither ends the run nor yields — it just wastes a turn and forces a retry. The lifecycle tools (``finish_scan`` for root, ``agent_finish`` for subagents) are the ONLY way to terminate, and the report flows through them. If you find yourself wanting to say "Done!" or "Scan complete" without a tool call, call the lifecycle tool instead.
+{%- if parallel_tool_calls %}
+- Read-only calls can go together in one turn: when several file reads or searches don't depend on each other's output, issue them as separate tool calls in the same turn rather than one per turn. Anything else — writing files, running commands that change state, sending requests, creating or messaging agents, reporting, or finishing — goes alone in its own turn.
+{%- endif %}
 {% endif %}
 </communication_rules>
 

strix/config/settings.py

@@ -65,6 +65,10 @@ class LlmSettings(BaseSettings):
         default=False,
         alias="STRIX_FORCE_REQUIRED_TOOL_CHOICE",
     )
+    parallel_tool_calls: bool = Field(
+        default=False,
+        alias="STRIX_PARALLEL_TOOL_CALLS",
+    )
     prompt_cache: bool = Field(
         default=True,
         alias="STRIX_PROMPT_CACHE",

strix/core/execution.py

@@ -1009,7 +1009,7 @@ async def _append_tool_required_message(
         message = (
             "Your previous message ended a turn without a tool call. Plain text never ends "
             "execution and never hands control to the user: it is shown to the user, and the "
-            "run continues. Continue immediately and call exactly one tool. "
+            "run continues. Continue immediately and call a tool. "
             "If you have nothing to do until the user replies, call wait_for_user; your "
             "text already reached them, so do not repeat it. "
             "If you are blocked waiting for another agent, call wait_for_agents. "
@@ -1021,7 +1021,7 @@ async def _append_tool_required_message(
         message = (
             "Your previous response ended the autonomous run without a lifecycle tool "
             "call. That is invalid in non-interactive mode; plain text final answers are "
-            "ignored. Continue immediately and call exactly one tool. "
+            "ignored. Continue immediately and call a tool. "
             f"If your work is complete, call {finish_tool}. "
             "If you are blocked waiting for another agent, call wait_for_agents. "
             "Otherwise use the appropriate execution or planning tool. "

strix/core/inputs.py

@@ -247,6 +247,7 @@ def make_model_settings(
     *,
     model_name: str,
     force_required_tool_choice: bool = False,
+    parallel_tool_calls: bool = False,
     request_timeout: float | None = None,
     prompt_cache: bool = True,
     extra_headers: dict[str, str] | None = None,
@@ -255,7 +256,7 @@ def make_model_settings(
 ) -> ModelSettings:
     headers = _request_headers(model_name, extra_headers)
     model_settings = ModelSettings(
-        parallel_tool_calls=False if has_tools else None,
+        parallel_tool_calls=parallel_tool_calls if has_tools else None,
         retry=DEFAULT_MODEL_RETRY,
         include_usage=True,
         extra_args=request_timeout_extra_args(request_timeout),

strix/core/runner.py

@@ -176,6 +176,7 @@ def _compose_root_instructions_override(
     is_whitebox: bool,
     is_diff_scoped: bool,
     interactive: bool,
+    parallel_tool_calls: bool,
     system_prompt_context: dict[str, Any],
     supports_images: bool,
     report_tool: str,
@@ -190,6 +191,7 @@ def _compose_root_instructions_override(
         is_root=True,
         is_diff_scoped=is_diff_scoped,
         interactive=interactive,
+        parallel_tool_calls=parallel_tool_calls,
         system_prompt_context=system_prompt_context,
         include_scope=False,
         supports_images=supports_images,
@@ -413,6 +415,7 @@ async def run_strix_scan(
             settings.llm.reasoning_effort,
             model_name=resolved_model,
             force_required_tool_choice=settings.llm.force_required_tool_choice,
+            parallel_tool_calls=settings.llm.parallel_tool_calls,
             request_timeout=settings.llm.timeout,
             prompt_cache=settings.llm.prompt_cache,
             extra_headers=settings.llm.extra_headers,
@@ -508,6 +511,7 @@ async def run_strix_scan(
             is_whitebox=is_whitebox,
             is_diff_scoped=is_diff_scoped,
             interactive=interactive,
+            parallel_tool_calls=settings.llm.parallel_tool_calls,
             system_prompt_context=root_context,
             supports_images=supports_images,
             report_tool=report_tool.name,
@@ -521,6 +525,7 @@ async def run_strix_scan(
             is_whitebox=is_whitebox,
             is_diff_scoped=is_diff_scoped,
             interactive=interactive,
+            parallel_tool_calls=settings.llm.parallel_tool_calls,
             chat_completions_tools=chat_completions_tools,
             strict_tool_schemas=strict_tool_schemas,
             system_prompt_context=root_context,
@@ -544,6 +549,7 @@ async def run_strix_scan(
             is_whitebox=is_whitebox,
             is_diff_scoped=is_diff_scoped,
             interactive=interactive,
+            parallel_tool_calls=settings.llm.parallel_tool_calls,
             chat_completions_tools=chat_completions_tools,
             strict_tool_schemas=strict_tool_schemas,
             system_prompt_context=scope_context,

tests/test_inputs.py

@@ -314,6 +314,12 @@ def test_make_model_settings_disables_parallel_tool_calls_by_default() -> None:
     assert make_model_settings("none", model_name="gpt-4o").parallel_tool_calls is False
 
 
+def test_make_model_settings_enables_parallel_tool_calls_when_requested() -> None:
+    settings = make_model_settings("none", model_name="gpt-4o", parallel_tool_calls=True)
+
+    assert settings.parallel_tool_calls is True
+
+
 def test_make_model_settings_omits_parallel_tool_calls_without_tools() -> None:
     settings = make_model_settings("none", model_name="gpt-4o", has_tools=False)
 

tests/test_runner_interrupt.py

@@ -25,6 +25,7 @@ def _wire_runner(monkeypatch: pytest.MonkeyPatch, tmp_path: Any) -> None:
             model="openai/gpt-4o",
             reasoning_effort="high",
             force_required_tool_choice=False,
+            parallel_tool_calls=False,
             timeout=300,
             prompt_cache=True,
             extra_headers=None,

tests/test_runner_mcp.py

@@ -41,6 +41,7 @@ def _settings() -> Any:
             model="openai/gpt-4o",
             reasoning_effort="high",
             force_required_tool_choice=False,
+            parallel_tool_calls=False,
             timeout=300,
             prompt_cache=True,
             extra_headers=None,

tests/test_runner_rate_limit.py

@@ -39,6 +39,7 @@ async def test_persistent_rate_limit_stops_gracefully(
             model="openai/gpt-4o",
             reasoning_effort="high",
             force_required_tool_choice=False,
+            parallel_tool_calls=False,
             timeout=300,
             prompt_cache=True,
             extra_headers=None,

tests/test_runner_root_prompt.py

@@ -61,6 +61,7 @@ def _patch_engine_scaffold(
             model="openai/gpt-4o",
             reasoning_effort="high",
             force_required_tool_choice=False,
+            parallel_tool_calls=False,
             timeout=300,
             prompt_cache=True,
             extra_headers=None,
@@ -285,6 +286,19 @@ async def test_unknown_tool_calls_are_returned_to_the_model(
     assert captured["run_config"].tool_not_found_behavior == "return_error_to_model"
 
 
+@pytest.mark.parametrize("interactive", [False, True])
+def test_parallel_tool_calls_prompt_is_opt_in(interactive: bool) -> None:
+    default = render_system_prompt(include_scope=False, interactive=interactive)
+    parallel = render_system_prompt(
+        include_scope=False, interactive=interactive, parallel_tool_calls=True
+    )
+
+    assert "Read-only calls can go together" not in default
+    assert ("Read-only calls can go together" in parallel) is not interactive
+    assert ("exactly one tool call" in default) is interactive
+    assert "exactly one tool call" not in parallel
+
+
 def test_scope_is_rendered_once_at_the_end_of_the_prompt() -> None:
     prompt = render_system_prompt(
         system_prompt_context={

tests/test_runner_teardown.py

@@ -49,6 +49,7 @@ def _settings() -> Any:
             model="openai/gpt-4o",
             reasoning_effort="high",
             force_required_tool_choice=False,
+            parallel_tool_calls=False,
             timeout=300,
             prompt_cache=True,
             extra_headers=None,