fix(subagents): consolidate system_prompt and skills into single SystemMessage (#2701)

* fix(subagents): consolidate system_prompt and skills into single SystemMessage

  Some LLM APIs (vLLM, Xinference, Chinese LLM providers) reject multiple
  system messages with \”System message must be at the beginning.\” The
  subagent executor was sending separate SystemMessages for the configured
  system_prompt and each loaded skill, which caused failures when calling
  task tool with sub-agents.

  Merge system_prompt and all skill content into one SystemMessage in the
  initial state, and pass system_prompt=None to create_agent() so the
  factory doesn't prepend a second one.

Fixes #2693

* fix(subagents): update SubagentConfig.system_prompt to str | None and add astream regression test

Agent-Logs-Url: https://github.com/bytedance/deer-flow/sessions/2ee03a26-e19b-4106-abc5-c76a2906383b

Co-authored-by: WillemJiang <219644+WillemJiang@users.noreply.github.com>

* fixed the lint error

* fix the lint error in the backend

* fix the unit test error of test_subagent_executor

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>

backend/packages/harness/deerflow/subagents/config.py

@@ -26,7 +26,7 @@ class SubagentConfig:
 
     name: str
     description: str
-    system_prompt: str
+    system_prompt: str | None = None
     tools: list[str] | None = None
     disallowed_tools: list[str] | None = field(default_factory=lambda: ["task"])
     skills: list[str] | None = None

backend/packages/harness/deerflow/subagents/executor.py

@@ -286,11 +286,13 @@ class SubagentExecutor:
         # Reuse shared middleware composition with lead agent.
         middlewares = build_subagent_runtime_middlewares(app_config=app_config, model_name=self.model_name, lazy_init=True)
 
+        # system_prompt is included in initial state messages (see _build_initial_state)
+        # to avoid multiple SystemMessages which some LLM APIs don't support.
         return create_agent(
             model=model,
             tools=tools if tools is not None else self.tools,
             middleware=middlewares,
-            system_prompt=self.config.system_prompt,
+            system_prompt=None,
             state_schema=ThreadState,
         )
 
@@ -365,14 +367,25 @@ class SubagentExecutor:
         Returns:
             Initial state dictionary and tools filtered by loaded skill metadata.
         """
+
         # Load skills as conversation items (Codex pattern)
         skills = await self._load_skills()
         filtered_tools = self._apply_skill_allowed_tools(skills)
         skill_messages = await self._load_skill_messages(skills)
 
+        # Combine system_prompt and skills into a single SystemMessage.
+        # Some LLM APIs reject multiple SystemMessages with
+        # "System message must be at the beginning."
+        system_parts: list[str] = []
+        if self.config.system_prompt:
+            system_parts.append(self.config.system_prompt)
+        for skill_msg in skill_messages:
+            system_parts.append(skill_msg.content)
+
         messages: list[Any] = []
-        # Skill content injected as developer/system messages before the task
-        messages.extend(skill_messages)
+        if system_parts:
+            messages.append(SystemMessage(content="\n\n".join(system_parts)))
+
         # Then the actual task
         messages.append(HumanMessage(content=task))