diff --git a/backend/app/agents/graph.py b/backend/app/agents/graph.py
index baffa5c..6f73bc7 100644
--- a/backend/app/agents/graph.py
+++ b/backend/app/agents/graph.py
@@ -36,12 +36,22 @@ from app.agents.runtime_metrics import (
 )
 from app.agents.schemas.event import AgentEvent
 from app.agents.schemas.message import AgentMessage
-from app.agents.schemas.task import AgentTask, CollaborationBudget, InterruptRecord, RecoveryRecord, TaskResult
+from app.agents.schemas.task import (
+    AgentTask,
+    CollaborationBudget,
+    InterruptRecord,
+    RecoveryRecord,
+    TaskResult,
+)
 from app.agents.skill_registry import build_skill_context
 from app.agents.state import AgentRole, AgentState
 from app.agents.tools import SUB_COMMANDER_TOOLSETS
 from app.agents.tools.time_reasoning import normalize_tool_time_arguments
-from app.agents.verifier import apply_verification_verdict, normalize_task_result, verify_task_result
+from app.agents.verifier import (
+    apply_verification_verdict,
+    normalize_task_result,
+    verify_task_result,
+)
 from app.services.llm_service import (
     create_llm_from_config,
     default_provider_capabilities,
@@ -67,13 +77,17 @@ ROLE_SKILL_CONTEXT = {
     AgentRole.EXECUTOR: "executor",
     AgentRole.LIBRARIAN: "librarian",
     AgentRole.ANALYST: "analyst",
+    AgentRole.CODE_COMMANDER: None,
 }
 
+from app.agents.prompts import CODE_COMMANDER_SYSTEM_PROMPT
+
 ROLE_SYSTEM_PROMPTS = {
     AgentRole.SCHEDULE_PLANNER: SCHEDULE_PLANNER_SYSTEM_PROMPT,
     AgentRole.EXECUTOR: EXECUTOR_SYSTEM_PROMPT,
     AgentRole.LIBRARIAN: LIBRARIAN_SYSTEM_PROMPT,
     AgentRole.ANALYST: ANALYST_SYSTEM_PROMPT,
+    AgentRole.CODE_COMMANDER: CODE_COMMANDER_SYSTEM_PROMPT,
 }
 
 SCHEDULE_KEYWORDS = (
@@ -105,7 +119,18 @@ ACCOUNTING_INTENT_KEYWORDS = (
     "花销",
     "开销",
 )
-KNOWLEDGE_KEYWORDS = ("知识", "搜索", "检索", "资料", "文档", "联网", "上网", "查询", "查一下", "最新")
+KNOWLEDGE_KEYWORDS = (
+    "知识",
+    "搜索",
+    "检索",
+    "资料",
+    "文档",
+    "联网",
+    "上网",
+    "查询",
+    "查一下",
+    "最新",
+)
 GENERAL_QA_PATTERNS = (
     "介绍一下",
     "介绍下",
@@ -118,8 +143,32 @@ GENERAL_QA_PATTERNS = (
 )
 ANALYSIS_KEYWORDS = ("分析", "报告", "统计", "趋势", "风险", "洞察", "总结")
 EXECUTION_KEYWORDS = ("创建", "更新", "修改", "执行", "发帖", "论坛", "帖子", "完成", "处理")
+CODE_COMMANDER_KEYWORDS = (
+    "写代码",
+    "代码",
+    "代码助手",
+    "demo",
+    "示例项目",
+    "贪食蛇",
+    "小游戏",
+    "帮我写一个",
+    "opencode",
+    "codex",
+    "claude code",
+    "gemini code",
+)
 SCHEDULE_ANALYSIS_KEYWORDS = ("聚焦", "判断", "分析", "优先级", "取舍", "最近对话", "该做什么")
-SCHEDULE_PLANNING_KEYWORDS = ("安排", "计划", "排期", "提醒", "创建", "新增", "会议", "交付", "节点")
+SCHEDULE_PLANNING_KEYWORDS = (
+    "安排",
+    "计划",
+    "排期",
+    "提醒",
+    "创建",
+    "新增",
+    "会议",
+    "交付",
+    "节点",
+)
 IDENTITY_PATTERNS = ("你是谁", "你是誰")
 CAPABILITY_PATTERNS = ("你能做什么", "你可以做什么", "你会做什么")
 SHORT_CONFIRMATION_PATTERNS = ("创建", "好的创建", "确认创建", "就创建", "那就创建")
@@ -130,13 +179,26 @@ SCHEDULE_CONFIRMATION_HINTS = (
     "是否需要我现在创建",
 )
 SCHEDULE_CONFIRMATION_QUESTION_MARKERS = ("是否", "要不要", "吗", "？", "?")
-COLLABORATION_STEP_MARKERS = ("然后", "再", "并且", "同时", "顺便", "最后", "分别", "拆成", "协作", "整合")
+COLLABORATION_STEP_MARKERS = (
+    "然后",
+    "再",
+    "并且",
+    "同时",
+    "顺便",
+    "最后",
+    "分别",
+    "拆成",
+    "协作",
+    "整合",
+)
 COLLABORATION_ROLE_ORDER = {
     AgentRole.LIBRARIAN: 0,
     AgentRole.ANALYST: 1,
     AgentRole.SCHEDULE_PLANNER: 2,
     AgentRole.EXECUTOR: 3,
 }
+
+
 def _role_value(role: AgentRole | str | None) -> str:
     if isinstance(role, AgentRole):
         return role.value
@@ -217,7 +279,9 @@ def _apply_isolation_payload(state: AgentState, payload: dict[str, Any]) -> None
     state["isolation_mode"] = str(payload.get("mode") or "none")
     state["isolation_id"] = str(payload.get("isolation_id") or "") or None
     state["isolation_workspace_path"] = str(payload.get("workspace_path") or "") or None
-    state["isolation_parent_conversation_id"] = str(payload.get("parent_conversation_id") or "") or None
+    state["isolation_parent_conversation_id"] = (
+        str(payload.get("parent_conversation_id") or "") or None
+    )
     state["isolation_metadata"] = dict(payload.get("metadata") or {})
 
 
@@ -325,8 +389,7 @@ def _record_response_usage(state: AgentState, response: Any) -> None:
 
     agent_id = str(state.get("agent_id") or state.get("current_agent") or AgentRole.MASTER.value)
     cost_by_agent = {
-        key: dict(value)
-        for key, value in dict(state.get("cost_by_agent") or {}).items()
+        key: dict(value) for key, value in dict(state.get("cost_by_agent") or {}).items()
     }
     agent_totals = dict(cost_by_agent.get(agent_id) or {})
     agent_input_tokens = int(agent_totals.get("input_tokens") or 0) + input_tokens
@@ -378,7 +441,9 @@ def _role_values() -> set[str]:
     return {role.value for role in AgentRole}
 
 
-def _summary_state_key(target: str) -> Literal["schedule_context_summary", "knowledge_context", "analysis_report"]:
+def _summary_state_key(
+    target: str,
+) -> Literal["schedule_context_summary", "knowledge_context", "analysis_report"]:
     if target not in {"schedule_context_summary", "knowledge_context", "analysis_report"}:
         raise ValueError(f"unsupported summary target: {target}")
     return cast(Literal["schedule_context_summary", "knowledge_context", "analysis_report"], target)
@@ -429,11 +494,15 @@ def _budget_snapshot_from_state(
         max_parallel_tasks=remaining_parallel_tasks,
         remaining_parallel_tasks=remaining_parallel_tasks,
         max_tool_calls=_get_state_int(state, "max_tool_rounds") or None,
-        remaining_tool_calls=max(_get_state_int(state, "max_tool_rounds") - _get_state_int(state, "tool_round_count"), 0)
+        remaining_tool_calls=max(
+            _get_state_int(state, "max_tool_rounds") - _get_state_int(state, "tool_round_count"), 0
+        )
         if _get_state_int(state, "max_tool_rounds")
         else None,
         max_iterations=_get_state_int(state, "max_iterations") or None,
-        remaining_iterations=max(_get_state_int(state, "max_iterations") - _get_state_int(state, "iteration_count"), 0)
+        remaining_iterations=max(
+            _get_state_int(state, "max_iterations") - _get_state_int(state, "iteration_count"), 0
+        )
         if _get_state_int(state, "max_iterations")
         else None,
         metadata=metadata_payload,
@@ -518,7 +587,11 @@ def _create_child_agent(
         _append_event_trace(
             state,
             "agent.spawn.blocked",
-            payload={"reason": "max_spawn_depth_exceeded", "target_role": role.value, "depth": current_depth},
+            payload={
+                "reason": "max_spawn_depth_exceeded",
+                "target_role": role.value,
+                "depth": current_depth,
+            },
             severity="warning",
             task_id=task.task_id,
             parent_task_id=task.parent_task_id,
@@ -542,7 +615,11 @@ def _create_child_agent(
     _append_event_trace(
         state,
         "agent.created",
-        payload={"parent_agent_id": parent_agent_id, "child_agent_id": child_agent_id, "target_role": role.value},
+        payload={
+            "parent_agent_id": parent_agent_id,
+            "child_agent_id": child_agent_id,
+            "target_role": role.value,
+        },
         task_id=task.task_id,
         parent_task_id=task.parent_task_id,
         child_task_id=task.task_id,
@@ -561,7 +638,9 @@ def _record_interrupt(
     task: AgentTask | dict[str, Any] | None = None,
     payload: dict[str, Any] | None = None,
 ) -> InterruptRecord:
-    task_payload = _task_payload(task) if task is not None else _collaboration_task_from_state(state)
+    task_payload = (
+        _task_payload(task) if task is not None else _collaboration_task_from_state(state)
+    )
     interrupt = InterruptRecord(
         interrupt_id=f"interrupt-{uuid4()}",
         reason=reason,
@@ -620,7 +699,9 @@ def _record_recovery(
     task: AgentTask | dict[str, Any] | None = None,
     payload: dict[str, Any] | None = None,
 ) -> RecoveryRecord:
-    task_payload = _task_payload(task) if task is not None else _collaboration_task_from_state(state)
+    task_payload = (
+        _task_payload(task) if task is not None else _collaboration_task_from_state(state)
+    )
     recovery = RecoveryRecord(
         recovery_id=f"recovery-{uuid4()}",
         source_interrupt_id=interrupt.interrupt_id,
@@ -696,7 +777,9 @@ def _resolve_capabilities(state: AgentState, llm) -> Any:
     capabilities = getattr(llm, "_jarvis_provider_capabilities", None)
     if capabilities is None:
         config = state.get("user_llm_config")
-        capabilities = resolve_provider_capabilities(config) if config else default_provider_capabilities()
+        capabilities = (
+            resolve_provider_capabilities(config) if config else default_provider_capabilities()
+        )
     state["provider_capabilities"] = {
         "provider": capabilities.provider,
         "supports_native_tools": capabilities.supports_native_tools,
@@ -728,7 +811,16 @@ def _coalesce_system_messages(messages: list[BaseMessage]) -> list[BaseMessage]:
 
 
 def _is_simple_greeting(text: str) -> bool:
-    return _normalize_user_text(text) in {"你好", "您好", "早", "早上好", "在吗", "嗨", "hi", "hello"}
+    return _normalize_user_text(text) in {
+        "你好",
+        "您好",
+        "早",
+        "早上好",
+        "在吗",
+        "嗨",
+        "hi",
+        "hello",
+    }
 
 
 def _is_identity_question(text: str) -> bool:
@@ -753,11 +845,7 @@ def _tool_result_indicates_failure(result: Any) -> bool:
 
 def _latest_assistant_message_content(messages: list[BaseMessage]) -> str:
     previous_assistant_message = next(
-        (
-            message
-            for message in reversed(messages[:-1])
-            if getattr(message, "type", "") == "ai"
-        ),
+        (message for message in reversed(messages[:-1]) if getattr(message, "type", "") == "ai"),
         None,
     )
     if previous_assistant_message is None:
@@ -770,7 +858,9 @@ def _previous_turn_proposed_schedule_creation(messages: list[BaseMessage]) -> bo
         return False
     content = _latest_assistant_message_content(messages)
     has_schedule_confirmation_hint = any(hint in content for hint in SCHEDULE_CONFIRMATION_HINTS)
-    has_question_marker = any(marker in content for marker in SCHEDULE_CONFIRMATION_QUESTION_MARKERS)
+    has_question_marker = any(
+        marker in content for marker in SCHEDULE_CONFIRMATION_QUESTION_MARKERS
+    )
     return has_schedule_confirmation_hint and has_question_marker
 
 
@@ -795,7 +885,9 @@ def _previous_turn_completed_reminder_creation(state: AgentState) -> bool:
     previous_message = messages[-2]
     if getattr(previous_message, "type", "") != "ai":
         return False
-    previous_assistant_content = _stringify_message_content(getattr(previous_message, "content", ""))
+    previous_assistant_content = _stringify_message_content(
+        getattr(previous_message, "content", "")
+    )
     completion_markers = ("已创建提醒", "提醒已经创建好了", "帮你设好了这条提醒", "创建成功")
     return any(marker in previous_assistant_content for marker in completion_markers)
 
@@ -822,7 +914,9 @@ def _expand_schedule_confirmation_query(user_query: str, messages: list[BaseMess
 def _is_schedule_creation_confirmation_response(response_text: str) -> bool:
     content = _stringify_message_content(response_text)
     has_schedule_confirmation_hint = any(hint in content for hint in SCHEDULE_CONFIRMATION_HINTS)
-    has_question_marker = any(marker in content for marker in SCHEDULE_CONFIRMATION_QUESTION_MARKERS)
+    has_question_marker = any(
+        marker in content for marker in SCHEDULE_CONFIRMATION_QUESTION_MARKERS
+    )
     return has_schedule_confirmation_hint and has_question_marker
 
 
@@ -848,7 +942,9 @@ def _clear_structured_continuity(state: AgentState) -> None:
     state["continuity_state"] = None
 
 
-def _should_clear_schedule_creation_continuity(state: AgentState, created_entities: list[dict[str, Any]]) -> bool:
+def _should_clear_schedule_creation_continuity(
+    state: AgentState, created_entities: list[dict[str, Any]]
+) -> bool:
     if not _has_active_structured_continuation(state):
         return False
     pending_action = state.get("pending_action") or {}
@@ -861,11 +957,14 @@ def _classify_request_signals(user_query: str) -> dict[str, Any]:
     text = (user_query or "").strip().lower()
     has_accounting_signal = any(keyword in text for keyword in ACCOUNTING_INTENT_KEYWORDS)
     has_schedule_signal = not has_accounting_signal and (
-        bool(re.search(r"\d{1,2}月\d{1,2}日", text)) or any(keyword in text for keyword in SCHEDULE_KEYWORDS)
+        bool(re.search(r"\d{1,2}月\d{1,2}日", text))
+        or any(keyword in text for keyword in SCHEDULE_KEYWORDS)
     )
     has_analysis_signal = any(keyword in text for keyword in ANALYSIS_KEYWORDS)
     has_knowledge_signal = any(keyword in text for keyword in KNOWLEDGE_KEYWORDS)
-    has_execution_signal = has_accounting_signal or any(keyword in text for keyword in EXECUTION_KEYWORDS)
+    has_execution_signal = has_accounting_signal or any(
+        keyword in text for keyword in EXECUTION_KEYWORDS
+    )
 
     roles: list[AgentRole] = []
     if has_knowledge_signal:
@@ -888,7 +987,9 @@ def _classify_request_signals(user_query: str) -> dict[str, Any]:
     }
 
 
-def _select_request_mode(user_query: str) -> tuple[Literal["direct", "collaboration"], dict[str, Any]]:
+def _select_request_mode(
+    user_query: str,
+) -> tuple[Literal["direct", "collaboration"], dict[str, Any]]:
     signals = _classify_request_signals(user_query)
     roles: list[AgentRole] = signals["roles"]
     text = signals["text"]
@@ -897,9 +998,12 @@ def _select_request_mode(user_query: str) -> tuple[Literal["direct", "collaborat
     is_long_request = len((user_query or "").strip()) >= 24
 
     should_collaborate = len(roles) >= 3 or (
-        len(roles) >= 2 and (has_multi_step_signal or is_explicit_collaboration_request or is_long_request)
+        len(roles) >= 2
+        and (has_multi_step_signal or is_explicit_collaboration_request or is_long_request)
+    )
+    selected_mode: Literal["direct", "collaboration"] = (
+        "collaboration" if should_collaborate else "direct"
     )
-    selected_mode: Literal["direct", "collaboration"] = "collaboration" if should_collaborate else "direct"
     metadata = {
         "mode": selected_mode,
         "reason": "multi_role_request" if should_collaborate else "single_role_or_simple_request",
@@ -913,6 +1017,10 @@ def _route_agent_from_user_query(user_query: str) -> AgentRole:
     signals = _classify_request_signals(user_query)
     text = signals["text"]
 
+    # Code Commander routing - check first as it has specific keywords
+    if any(keyword in text for keyword in CODE_COMMANDER_KEYWORDS):
+        return AgentRole.CODE_COMMANDER
+
     if signals["has_accounting_signal"]:
         return AgentRole.EXECUTOR
 
@@ -937,7 +1045,9 @@ def _choose_sub_commander(role: AgentRole, user_query: str) -> str:
     text = (user_query or "").strip().lower()
 
     if role == AgentRole.SCHEDULE_PLANNER:
-        if re.search(r"\d{1,2}月\d{1,2}日", text) or any(keyword in text for keyword in SCHEDULE_PLANNING_KEYWORDS):
+        if re.search(r"\d{1,2}月\d{1,2}日", text) or any(
+            keyword in text for keyword in SCHEDULE_PLANNING_KEYWORDS
+        ):
             return "schedule_planning"
         return "schedule_analysis"
     if role == AgentRole.EXECUTOR:
@@ -965,7 +1075,6 @@ def _is_missing_knowledge_result(tool_result: str | None) -> bool:
         "未找到相关网页结果",
         "暂无相关记录",
         "没有找到",
-
     )
     return any(marker in text for marker in markers)
 
@@ -985,7 +1094,7 @@ def _extract_json_object(content: str) -> str | None:
             continue
         try:
             _, end = decoder.raw_decode(text[index:])
-            return text[index:index + end]
+            return text[index : index + end]
         except json.JSONDecodeError:
             continue
     return None
@@ -1096,7 +1205,9 @@ def _build_structured_continuity_summary(state: AgentState) -> str | None:
     return "\n".join(lines)
 
 
-def _build_system_messages(state: AgentState, system_prompt: str, role: AgentRole, sub_commander: str) -> list[BaseMessage]:
+def _build_system_messages(
+    state: AgentState, system_prompt: str, role: AgentRole, sub_commander: str
+) -> list[BaseMessage]:
     messages: list[BaseMessage] = [SystemMessage(content=system_prompt)]
 
     current_datetime_context = state.get("current_datetime_context")
@@ -1164,10 +1275,16 @@ def _build_task_goal(role: AgentRole, user_query: str) -> str:
 
 def _build_expected_evidence(role: AgentRole) -> list[dict[str, Any]]:
     evidence_map = {
-        AgentRole.LIBRARIAN: [{"type": "evidence", "detail": "retrieval findings or source-backed context"}],
+        AgentRole.LIBRARIAN: [
+            {"type": "evidence", "detail": "retrieval findings or source-backed context"}
+        ],
         AgentRole.ANALYST: [{"type": "analysis", "detail": "judgment with supporting rationale"}],
-        AgentRole.SCHEDULE_PLANNER: [{"type": "plan", "detail": "explicit schedule or next-step proposal"}],
-        AgentRole.EXECUTOR: [{"type": "execution", "detail": "tool output or execution confirmation"}],
+        AgentRole.SCHEDULE_PLANNER: [
+            {"type": "plan", "detail": "explicit schedule or next-step proposal"}
+        ],
+        AgentRole.EXECUTOR: [
+            {"type": "execution", "detail": "tool output or execution confirmation"}
+        ],
     }
     return evidence_map.get(role, [{"type": "summary", "detail": "task evidence"}])
 
@@ -1185,7 +1302,9 @@ def _build_collaboration_tasks(user_query: str) -> list[AgentTask]:
             unique_roles.append(role)
 
     parent_task_id = f"task-root-{uuid4().hex[:8]}"
-    task_ids = [f"task-{index}-{uuid4().hex[:8]}" for index, _ in enumerate(unique_roles[:4], start=1)]
+    task_ids = [
+        f"task-{index}-{uuid4().hex[:8]}" for index, _ in enumerate(unique_roles[:4], start=1)
+    ]
     tasks: list[AgentTask] = []
     for index, role in enumerate(unique_roles[:4]):
         tasks.append(
@@ -1224,7 +1343,9 @@ def _build_collaboration_context_summary(state: AgentState) -> str | None:
     if task_results:
         lines.append("completed_collaboration_results:")
         for item in task_results[-4:]:
-            normalized = item.model_dump(mode="json") if isinstance(item, TaskResult) else dict(item)
+            normalized = (
+                item.model_dump(mode="json") if isinstance(item, TaskResult) else dict(item)
+            )
             lines.append(
                 f"- {normalized.get('task_id')} | {normalized.get('status')} | {normalized.get('summary') or ''}"
             )
@@ -1247,7 +1368,9 @@ def _maybe_reset_turn_budgets(state: AgentState) -> None:
         return
 
     last_message_type = getattr(messages[-1], "type", "")
-    has_prior_assistant_turn = any(getattr(message, "type", "") == "ai" for message in messages[:-1])
+    has_prior_assistant_turn = any(
+        getattr(message, "type", "") == "ai" for message in messages[:-1]
+    )
     if last_message_type in {"human", "user"} and has_prior_assistant_turn:
         state["routing_hops"] = 0
         state["terminated_due_to_loop_guard"] = False
@@ -1301,7 +1424,9 @@ def _append_event_trace(
     ]
 
 
-def _set_phase(state: AgentState, phase: str, *, reason: str, payload: dict[str, Any] | None = None) -> None:
+def _set_phase(
+    state: AgentState, phase: str, *, reason: str, payload: dict[str, Any] | None = None
+) -> None:
     if state.get("current_phase") == phase:
         return
     state["current_phase"] = phase
@@ -1320,7 +1445,9 @@ def _set_phase(state: AgentState, phase: str, *, reason: str, payload: dict[str,
     )
 
 
-def _record_checkpoint(state: AgentState, checkpoint: str, *, reason: str, payload: dict[str, Any] | None = None) -> None:
+def _record_checkpoint(
+    state: AgentState, checkpoint: str, *, reason: str, payload: dict[str, Any] | None = None
+) -> None:
     state["current_checkpoint"] = checkpoint
     state["checkpoint_history"] = [
         *(state.get("checkpoint_history") or []),
@@ -1334,7 +1461,12 @@ def _record_checkpoint(state: AgentState, checkpoint: str, *, reason: str, paylo
     _append_event_trace(
         state,
         "agent.checkpoint.recorded",
-        payload={"checkpoint": checkpoint, "phase": state.get("current_phase"), "reason": reason, **(payload or {})},
+        payload={
+            "checkpoint": checkpoint,
+            "phase": state.get("current_phase"),
+            "reason": reason,
+            **(payload or {}),
+        },
     )
 
 
@@ -1384,7 +1516,9 @@ def _update_task_result_summary(state: AgentState, tool_summaries: list[dict[str
     state["action_results"] = [*(state.get("action_results") or []), summary]
 
 
-def _record_sub_commander(state: AgentState, role: AgentRole, sub_commander: str, user_query: str) -> None:
+def _record_sub_commander(
+    state: AgentState, role: AgentRole, sub_commander: str, user_query: str
+) -> None:
     thread_id = _ensure_message_thread(state)
     message_id, message_index = _bump_message_sequence(state)
     current_agent_id = str(state.get("agent_id") or AgentRole.MASTER.value)
@@ -1428,10 +1562,14 @@ def _record_sub_commander(state: AgentState, role: AgentRole, sub_commander: str
 
 def _stop_sub_commander_due_to_budget(state: AgentState, reason: str) -> None:
     state["stop_reason"] = reason
-    state["final_response"] = "这次需要处理的步骤有点多，我先停在这里。您可以把目标再明确一点，或让我先只完成其中一步。"
+    state["final_response"] = (
+        "这次需要处理的步骤有点多，我先停在这里。您可以把目标再明确一点，或让我先只完成其中一步。"
+    )
 
 
-def _guard_sub_commander_budget(state: AgentState, counter_key: str, max_key: str, reason: str) -> bool:
+def _guard_sub_commander_budget(
+    state: AgentState, counter_key: str, max_key: str, reason: str
+) -> bool:
     max_value = _get_state_int(state, max_key)
     current_value = _get_state_int(state, counter_key)
     if max_value > 0 and current_value >= max_value:
@@ -1533,7 +1671,9 @@ def _build_clarification_summary(state: AgentState) -> str | None:
         lines.append(f"- missing_fields: {', '.join(str(field) for field in missing_fields)}")
     if partial_args:
         lines.append(f"- partial_args: {json.dumps(partial_args, ensure_ascii=False)}")
-    lines.append("- instruction: merge the user's latest answer into the missing fields and continue the same action")
+    lines.append(
+        "- instruction: merge the user's latest answer into the missing fields and continue the same action"
+    )
     return "\n".join(lines)
 
 
@@ -1593,9 +1733,13 @@ def _canonicalize_tool_call(tool_name: str, args: dict[str, Any]) -> tuple[str,
     normalized_args = dict(args)
 
     if normalized_name == "create_reminder":
-        if normalized_args.get("description") and not (normalized_args.get("title") or normalized_args.get("content")):
+        if normalized_args.get("description") and not (
+            normalized_args.get("title") or normalized_args.get("content")
+        ):
             normalized_args["content"] = normalized_args["description"]
-        if normalized_args.get("reminder_content") and not (normalized_args.get("title") or normalized_args.get("content")):
+        if normalized_args.get("reminder_content") and not (
+            normalized_args.get("title") or normalized_args.get("content")
+        ):
             normalized_args["content"] = normalized_args["reminder_content"]
         if normalized_args.get("reminder_time") and not (
             normalized_args.get("reminder_at")
@@ -1607,15 +1751,21 @@ def _canonicalize_tool_call(tool_name: str, args: dict[str, Any]) -> tuple[str,
             normalized_args["time"] = normalized_args["reminder_time"]
 
     if normalized_name in {"create_schedule_task", "create_task"}:
-        if normalized_args.get("task") and not (normalized_args.get("title") or normalized_args.get("content")):
+        if normalized_args.get("task") and not (
+            normalized_args.get("title") or normalized_args.get("content")
+        ):
             normalized_args["title"] = normalized_args["task"]
-        if normalized_args.get("due_datetime") and not (normalized_args.get("due_date") or normalized_args.get("date")):
+        if normalized_args.get("due_datetime") and not (
+            normalized_args.get("due_date") or normalized_args.get("date")
+        ):
             normalized_args["due_date"] = normalized_args["due_datetime"]
         if normalized_args.get("due_time") and not normalized_args.get("due_date"):
             normalized_args["due_date"] = normalized_args["due_time"]
 
     if normalized_name == "create_todo":
-        if normalized_args.get("task") and not (normalized_args.get("title") or normalized_args.get("content")):
+        if normalized_args.get("task") and not (
+            normalized_args.get("title") or normalized_args.get("content")
+        ):
             normalized_args["title"] = normalized_args["task"]
         if normalized_args.get("date") and not normalized_args.get("todo_date"):
             normalized_args["todo_date"] = normalized_args["date"]
@@ -1623,7 +1773,10 @@ def _canonicalize_tool_call(tool_name: str, args: dict[str, Any]) -> tuple[str,
             normalized_args["todo_date"] = normalized_args["due_date"]
         if normalized_args.get("due_datetime") and not normalized_args.get("todo_date"):
             normalized_args["todo_date"] = normalized_args["due_datetime"]
-        if any(normalized_args.get(key) for key in ("due_datetime", "start_time", "end_time", "due_time")):
+        if any(
+            normalized_args.get(key)
+            for key in ("due_datetime", "start_time", "end_time", "due_time")
+        ):
             normalized_name = "create_schedule_task"
             if normalized_args.get("due_time") and not normalized_args.get("due_date"):
                 normalized_args["due_date"] = normalized_args["due_time"]
@@ -1631,7 +1784,9 @@ def _canonicalize_tool_call(tool_name: str, args: dict[str, Any]) -> tuple[str,
                 normalized_args["due_date"] = normalized_args["todo_date"]
 
     if normalized_name == "create_goal":
-        if normalized_args.get("task") and not (normalized_args.get("title") or normalized_args.get("content")):
+        if normalized_args.get("task") and not (
+            normalized_args.get("title") or normalized_args.get("content")
+        ):
             normalized_args["title"] = normalized_args["task"]
         if normalized_args.get("date") and not normalized_args.get("goal_date"):
             normalized_args["goal_date"] = normalized_args["date"]
@@ -1740,7 +1895,9 @@ async def _execute_tool_calls(
             {
                 "tool_name": tool_name,
                 "result_preview": _stringify_message_content(result)[:200],
-                "created_entity_types": [entity.get("type") for entity in call_created_entities if entity.get("type")],
+                "created_entity_types": [
+                    entity.get("type") for entity in call_created_entities if entity.get("type")
+                ],
                 "created_count": len(call_created_entities),
             }
         )
@@ -1765,7 +1922,9 @@ async def _run_sub_commander(
 
     llm = _get_llm_for_state(state)
     capabilities = _resolve_capabilities(state, llm)
-    if _has_active_clarification_context(state) and role.value == str((state.get("clarification_context") or {}).get("owning_agent") or ""):
+    if _has_active_clarification_context(state) and role.value == str(
+        (state.get("clarification_context") or {}).get("owning_agent") or ""
+    ):
         user_query = _build_resumed_clarification_query(state, user_query)
     sub_commander = _choose_sub_commander(role, user_query)
     _record_sub_commander(state, role, sub_commander, user_query)
@@ -1785,7 +1944,9 @@ async def _run_sub_commander(
     ):
         state["tool_calls"] = []
         state["last_tool_result"] = None
-        state["final_response"] = "上一条提醒已经创建好了。若您现在要新建别的内容，请直接告诉我要创建什么。"
+        state["final_response"] = (
+            "上一条提醒已经创建好了。若您现在要新建别的内容，请直接告诉我要创建什么。"
+        )
         history_messages = list(state.get("messages", []))
         history_messages.append(AIMessage(content=state["final_response"]))
         state["messages"] = history_messages
@@ -1804,12 +1965,18 @@ async def _run_sub_commander(
     state["fallback_parse_error"] = None
     start_tool_index = len(state.get("tool_outcomes") or [])
 
-    if not _guard_sub_commander_budget(state, "tool_round_count", "max_tool_rounds", "max_tool_rounds_exceeded"):
+    if not _guard_sub_commander_budget(
+        state, "tool_round_count", "max_tool_rounds", "max_tool_rounds_exceeded"
+    ):
         pass
-    elif not _guard_sub_commander_budget(state, "retry_count", "max_retries", "max_retries_exceeded"):
+    elif not _guard_sub_commander_budget(
+        state, "retry_count", "max_retries", "max_retries_exceeded"
+    ):
         pass
     elif not toolset:
-        if _guard_sub_commander_budget(state, "iteration_count", "max_iterations", "max_iterations_exceeded"):
+        if _guard_sub_commander_budget(
+            state, "iteration_count", "max_iterations", "max_iterations_exceeded"
+        ):
             state["iteration_count"] = int(state.get("iteration_count") or 0) + 1
             response = await _invoke_llm(llm, working_messages)
             _record_response_usage(state, response)
@@ -1818,14 +1985,18 @@ async def _run_sub_commander(
         state["tool_strategy_used"] = "native"
         bound_llm = llm.bind_tools(toolset)
         while state.get("final_response") is None and not state.get("clarification_needed"):
-            if not _guard_sub_commander_budget(state, "iteration_count", "max_iterations", "max_iterations_exceeded"):
+            if not _guard_sub_commander_budget(
+                state, "iteration_count", "max_iterations", "max_iterations_exceeded"
+            ):
                 break
             state["iteration_count"] = int(state.get("iteration_count") or 0) + 1
             response = await _invoke_llm(bound_llm, working_messages)
             _record_response_usage(state, response)
             tool_calls = getattr(response, "tool_calls", None) or []
             if tool_calls:
-                if not _guard_sub_commander_budget(state, "tool_round_count", "max_tool_rounds", "max_tool_rounds_exceeded"):
+                if not _guard_sub_commander_budget(
+                    state, "tool_round_count", "max_tool_rounds", "max_tool_rounds_exceeded"
+                ):
                     break
                 prepared_calls, clarification = _prepare_tool_calls_for_execution(tool_calls, state)
                 if clarification:
@@ -1848,28 +2019,47 @@ async def _run_sub_commander(
                     content=_stringify_message_content(getattr(response, "content", "")),
                     tool_calls=tool_calls,
                 )
-                normalized_calls, tool_result, created_entities, tool_messages = await _execute_tool_calls(
+                (
+                    normalized_calls,
+                    tool_result,
+                    created_entities,
+                    tool_messages,
+                ) = await _execute_tool_calls(
                     prepared_calls,
                     toolset,
                     state,
                 )
                 state["tool_calls"] = normalized_calls
                 state["last_tool_result"] = tool_result
-                state["created_entities"] = [*(state.get("created_entities") or []), *created_entities]
+                state["created_entities"] = [
+                    *(state.get("created_entities") or []),
+                    *created_entities,
+                ]
                 if created_entities:
                     _clear_clarification_context(state)
-                if role == AgentRole.SCHEDULE_PLANNER and _should_clear_schedule_creation_continuity(state, created_entities):
+                if (
+                    role == AgentRole.SCHEDULE_PLANNER
+                    and _should_clear_schedule_creation_continuity(state, created_entities)
+                ):
                     _clear_structured_continuity(state)
                 working_messages = [*working_messages, assistant_tool_message, *tool_messages]
-                if sub_commander == "librarian_retrieval" and _is_missing_knowledge_result(tool_result):
-                    working_messages.append(SystemMessage(content="如果检索工具没有找到证据，可以直接基于你的常识给出清晰回答，不要机械地说不知道。"))
+                if sub_commander == "librarian_retrieval" and _is_missing_knowledge_result(
+                    tool_result
+                ):
+                    working_messages.append(
+                        SystemMessage(
+                            content="如果检索工具没有找到证据，可以直接基于你的常识给出清晰回答，不要机械地说不知道。"
+                        )
+                    )
                 continue
             state["final_response"] = _stringify_message_content(response.content)
     else:
         state["tool_strategy_used"] = "json_fallback"
         allowed_tools = [tool.name for tool in toolset]
         while state.get("final_response") is None and not state.get("clarification_needed"):
-            if not _guard_sub_commander_budget(state, "iteration_count", "max_iterations", "max_iterations_exceeded"):
+            if not _guard_sub_commander_budget(
+                state, "iteration_count", "max_iterations", "max_iterations_exceeded"
+            ):
                 break
             state["iteration_count"] = int(state.get("iteration_count") or 0) + 1
             parsed = None
@@ -1894,16 +2084,22 @@ async def _run_sub_commander(
                 if parsed is not None:
                     state["fallback_parse_error"] = None
                     break
-                if not _guard_sub_commander_budget(state, "iteration_count", "max_iterations", "max_iterations_exceeded"):
+                if not _guard_sub_commander_budget(
+                    state, "iteration_count", "max_iterations", "max_iterations_exceeded"
+                ):
                     parsed = None
                     break
                 if int(state.get("retry_count") or 0) >= int(state.get("max_retries") or 0):
                     state["fallback_parse_error"] = "invalid_json_action"
-                    state["final_response"] = "这次内部动作解析没整理好，不过您的意思我接住了。您再说一遍要我执行的内容，我只回结果，不展示内部调用细节。"
+                    state["final_response"] = (
+                        "这次内部动作解析没整理好，不过您的意思我接住了。您再说一遍要我执行的内容，我只回结果，不展示内部调用细节。"
+                    )
                     break
                 state["iteration_count"] = int(state.get("iteration_count") or 0) + 1
                 state["retry_count"] = int(state.get("retry_count") or 0) + 1
-                retry_instruction = SystemMessage(content="上一次输出不是有效 JSON。请严格只返回合法 JSON，不要加解释。")
+                retry_instruction = SystemMessage(
+                    content="上一次输出不是有效 JSON。请严格只返回合法 JSON，不要加解释。"
+                )
             if state.get("final_response") is not None:
                 break
             if parsed is None:
@@ -1924,9 +2120,13 @@ async def _run_sub_commander(
                 state["stop_reason"] = "clarification_needed"
                 state["final_response"] = parsed["clarification_question"]
                 break
-            if not _guard_sub_commander_budget(state, "tool_round_count", "max_tool_rounds", "max_tool_rounds_exceeded"):
+            if not _guard_sub_commander_budget(
+                state, "tool_round_count", "max_tool_rounds", "max_tool_rounds_exceeded"
+            ):
                 break
-            prepared_calls, clarification = _prepare_tool_calls_for_execution(parsed["tool_calls"], state)
+            prepared_calls, clarification = _prepare_tool_calls_for_execution(
+                parsed["tool_calls"], state
+            )
             if clarification:
                 state["clarification_needed"] = True
                 state["clarification_question"] = clarification["question"]
@@ -1943,7 +2143,12 @@ async def _run_sub_commander(
                 state["final_response"] = clarification["question"]
                 break
             state["tool_round_count"] = int(state.get("tool_round_count") or 0) + 1
-            normalized_calls, tool_result, created_entities, tool_messages = await _execute_tool_calls(
+            (
+                normalized_calls,
+                tool_result,
+                created_entities,
+                tool_messages,
+            ) = await _execute_tool_calls(
                 prepared_calls,
                 toolset,
                 state,
@@ -1951,11 +2156,17 @@ async def _run_sub_commander(
             state["tool_calls"] = normalized_calls
             state["last_tool_result"] = tool_result
             state["created_entities"] = [*(state.get("created_entities") or []), *created_entities]
-            if role == AgentRole.SCHEDULE_PLANNER and _should_clear_schedule_creation_continuity(state, created_entities):
+            if role == AgentRole.SCHEDULE_PLANNER and _should_clear_schedule_creation_continuity(
+                state, created_entities
+            ):
                 _clear_structured_continuity(state)
             working_messages = [*working_messages, *tool_messages]
             if sub_commander == "librarian_retrieval" and _is_missing_knowledge_result(tool_result):
-                working_messages.append(SystemMessage(content="如果检索工具没有找到证据，可以直接基于你的常识给出清晰回答，不要机械地说不知道。"))
+                working_messages.append(
+                    SystemMessage(
+                        content="如果检索工具没有找到证据，可以直接基于你的常识给出清晰回答，不要机械地说不知道。"
+                    )
+                )
 
     if summary_target:
         state[_summary_state_key(summary_target)] = state.get("final_response")
@@ -1963,8 +2174,7 @@ async def _run_sub_commander(
     task_result_summary = state.get("task_result_summary")
     tool_outcomes = list(state.get("tool_outcomes") or [])[start_tool_index:]
     has_tool_failure = any(
-        _tool_result_indicates_failure(outcome.get("result_preview"))
-        for outcome in tool_outcomes
+        _tool_result_indicates_failure(outcome.get("result_preview")) for outcome in tool_outcomes
     )
     verifier_input = {
         "summary": state.get("final_response") or (task_result_summary or {}).get("tools"),
@@ -2083,7 +2293,9 @@ def _collect_task_result(task: AgentTask, state: AgentState, start_tool_index: i
     return TaskResult(
         task_id=task.task_id,
         status=status,
-        summary=_stringify_message_content(state.get("final_response") or state.get("verification_summary")),
+        summary=_stringify_message_content(
+            state.get("final_response") or state.get("verification_summary")
+        ),
         evidence=_build_task_evidence(state, start_tool_index),
         owner_agent_id=task.owner_agent_id,
         parent_task_id=task.parent_task_id,
@@ -2107,7 +2319,9 @@ def _collect_task_result(task: AgentTask, state: AgentState, start_tool_index: i
     )
 
 
-def _apply_task_result_to_state(state: AgentState, task: AgentTask, task_result: TaskResult) -> None:
+def _apply_task_result_to_state(
+    state: AgentState, task: AgentTask, task_result: TaskResult
+) -> None:
     normalized_result = normalize_task_result(task_result, default_task_id=task.task_id)
     serialized_result = normalized_result.model_dump(mode="json")
     state["task_results"] = [*(state.get("task_results") or []), serialized_result]
@@ -2116,7 +2330,11 @@ def _apply_task_result_to_state(state: AgentState, task: AgentTask, task_result:
     completed_entry: dict[str, Any] | None = None
     pending_tasks: list[dict[str, Any]] = []
     for existing_task in state.get("active_tasks") or []:
-        normalized_task = existing_task.model_dump(mode="json") if isinstance(existing_task, AgentTask) else dict(existing_task)
+        normalized_task = (
+            existing_task.model_dump(mode="json")
+            if isinstance(existing_task, AgentTask)
+            else dict(existing_task)
+        )
         if normalized_task.get("task_id") == task.task_id:
             normalized_task["status"] = normalized_result.status
             normalized_task["evidence"] = list(normalized_result.evidence)
@@ -2142,10 +2360,7 @@ def _apply_task_result_to_state(state: AgentState, task: AgentTask, task_result:
 
 
 def _build_collaboration_final_response(task_results: list[TaskResult | dict[str, Any]]) -> str:
-    normalized_results = [
-        normalize_task_result(item)
-        for item in task_results
-    ]
+    normalized_results = [normalize_task_result(item) for item in task_results]
     lines = [f"已按协作模式回收 {len(normalized_results)} 个子任务结果："]
     final_completed_summary: str | None = None
     for index, result in enumerate(normalized_results, start=1):
@@ -2157,7 +2372,9 @@ def _build_collaboration_final_response(task_results: list[TaskResult | dict[str
     if final_completed_summary:
         lines.append(f"汇总结论：{final_completed_summary}")
     elif normalized_results:
-        blocked_result = next((item for item in reversed(normalized_results) if item.next_action), None)
+        blocked_result = next(
+            (item for item in reversed(normalized_results) if item.next_action), None
+        )
         if blocked_result and blocked_result.next_action:
             lines.append(f"下一步：{blocked_result.next_action}")
         else:
@@ -2173,21 +2390,17 @@ def _verify_collaboration_results(
     expected_task_ids = {task.task_id for task in tasks}
     normalized_results = [
         normalize_task_result(item)
-        for item in (task_results if task_results is not None else (state.get("task_results") or []))
+        for item in (
+            task_results if task_results is not None else (state.get("task_results") or [])
+        )
         if normalize_task_result(item).task_id in expected_task_ids
     ]
     result_by_task_id = {item.task_id: item for item in normalized_results}
     missing_task_ids = [task.task_id for task in tasks if task.task_id not in result_by_task_id]
     failed_or_blocked = [
-        item.task_id
-        for item in normalized_results
-        if item.status in {"failed", "blocked"}
-    ]
-    missing_evidence = [
-        item.task_id
-        for item in normalized_results
-        if not item.evidence
+        item.task_id for item in normalized_results if item.status in {"failed", "blocked"}
     ]
+    missing_evidence = [item.task_id for item in normalized_results if not item.evidence]
 
     verification_evidence = [
         {
@@ -2200,16 +2413,24 @@ def _verify_collaboration_results(
     ]
     if missing_task_ids:
         summary = f"协作结果不完整，缺少任务结果: {', '.join(missing_task_ids)}"
-        verdict = verify_task_result(status="failed", summary=summary, evidence=verification_evidence)
+        verdict = verify_task_result(
+            status="failed", summary=summary, evidence=verification_evidence
+        )
     elif failed_or_blocked:
         summary = f"协作结果未闭环，存在失败或阻塞任务: {', '.join(failed_or_blocked)}"
-        verdict = verify_task_result(status="failed", summary=summary, evidence=verification_evidence)
+        verdict = verify_task_result(
+            status="failed", summary=summary, evidence=verification_evidence
+        )
     elif missing_evidence:
         summary = f"协作结果证据不足，缺少 evidence 的任务: {', '.join(missing_evidence)}"
-        verdict = verify_task_result(status="failed", summary=summary, evidence=verification_evidence)
+        verdict = verify_task_result(
+            status="failed", summary=summary, evidence=verification_evidence
+        )
     else:
         summary = f"协作模式已完成 {len(normalized_results)}/{len(tasks)} 个子任务，并为每个子任务回收了结果与 evidence。"
-        verdict = verify_task_result(status="passed", summary=summary, evidence=verification_evidence)
+        verdict = verify_task_result(
+            status="passed", summary=summary, evidence=verification_evidence
+        )
 
     updated_state = apply_verification_verdict(state, verdict)
     state.update(updated_state)
@@ -2222,13 +2443,30 @@ async def _run_collaboration_flow(state: AgentState, user_query: str) -> AgentSt
     if len(tasks) < 2:
         state["execution_mode"] = "direct"
         state["routing_decision"] = {"mode": "direct", "reason": "collaboration_plan_fell_back"}
-        _record_checkpoint(state, "collaboration.fallback_to_direct", reason="insufficient_tasks", payload={"task_count": len(tasks)})
-        _set_phase(state, "phase_1_routing", reason="collaboration_flow_abandoned", payload={"task_count": len(tasks)})
-        _record_checkpoint(state, "routing.direct_resumed", reason="collaboration_flow_abandoned", payload={"task_count": len(tasks)})
+        _record_checkpoint(
+            state,
+            "collaboration.fallback_to_direct",
+            reason="insufficient_tasks",
+            payload={"task_count": len(tasks)},
+        )
+        _set_phase(
+            state,
+            "phase_1_routing",
+            reason="collaboration_flow_abandoned",
+            payload={"task_count": len(tasks)},
+        )
+        _record_checkpoint(
+            state,
+            "routing.direct_resumed",
+            reason="collaboration_flow_abandoned",
+            payload={"task_count": len(tasks)},
+        )
         return state
 
     base_history = list(state.get("messages", []))
-    root_agent_id = str(state.get("root_agent_id") or state.get("agent_id") or AgentRole.MASTER.value)
+    root_agent_id = str(
+        state.get("root_agent_id") or state.get("agent_id") or AgentRole.MASTER.value
+    )
     coordinator_agent_id = str(state.get("agent_id") or AgentRole.MASTER.value)
     state["execution_mode"] = "collaboration"
     state["routing_decision"] = {
@@ -2250,15 +2488,24 @@ async def _run_collaboration_flow(state: AgentState, user_query: str) -> AgentSt
         payload=budget_snapshot,
     )
     state["active_tasks"] = [task.model_dump(mode="json") for task in tasks]
-    _record_checkpoint(state, "collaboration.tasks_ready", reason="tasks_built", payload={"task_count": len(tasks)})
-    parent_task_id = next((task.parent_task_id for task in tasks if task.parent_task_id), None) or "root"
+    _record_checkpoint(
+        state, "collaboration.tasks_ready", reason="tasks_built", payload={"task_count": len(tasks)}
+    )
+    parent_task_id = (
+        next((task.parent_task_id for task in tasks if task.parent_task_id), None) or "root"
+    )
     state["task_hierarchy"] = {parent_task_id: [task.task_id for task in tasks]}
     state["task_results"] = []
     state["next_step"] = None
 
     _set_phase(state, "phase_3_dynamic_collaboration", reason="collaboration_workers_dispatch")
     for task in tasks:
-        _record_checkpoint(state, "collaboration.task_dispatch", reason="dispatch_task", payload={"task_id": task.task_id, "role": task.role})
+        _record_checkpoint(
+            state,
+            "collaboration.task_dispatch",
+            reason="dispatch_task",
+            payload={"task_id": task.task_id, "role": task.role},
+        )
         state["current_agent"] = AgentRole.MASTER.value
         state["agent_id"] = coordinator_agent_id
         state["parent_agent_id"] = None
@@ -2311,7 +2558,12 @@ async def _run_collaboration_flow(state: AgentState, user_query: str) -> AgentSt
         )
 
         task_result = _collect_task_result(task, state, start_tool_index)
-        _record_checkpoint(state, "collaboration.task_result_collected", reason="task_finished", payload={"task_id": task.task_id, "status": task_result.status})
+        _record_checkpoint(
+            state,
+            "collaboration.task_result_collected",
+            reason="task_finished",
+            payload={"task_id": task.task_id, "status": task_result.status},
+        )
         _append_message_trace(
             state,
             from_agent_id=child_agent_id,
@@ -2343,7 +2595,9 @@ async def _run_collaboration_flow(state: AgentState, user_query: str) -> AgentSt
     state["root_agent_id"] = root_agent_id
     state["collaboration_depth"] = 0
     state["final_response"] = _build_collaboration_final_response(state.get("task_results") or [])
-    _set_phase(state, "phase_4_visibility_and_verification", reason="collaboration_verification_started")
+    _set_phase(
+        state, "phase_4_visibility_and_verification", reason="collaboration_verification_started"
+    )
     _record_checkpoint(state, "collaboration.verification_started", reason="before_verify")
     _append_event_trace(
         state,
@@ -2364,7 +2618,12 @@ async def _run_collaboration_flow(state: AgentState, user_query: str) -> AgentSt
         },
         severity="error" if state.get("verification_status") == "failed" else "info",
     )
-    _record_checkpoint(state, "collaboration.completed", reason="collaboration_flow_finished", payload={"verification_status": state.get("verification_status")})
+    _record_checkpoint(
+        state,
+        "collaboration.completed",
+        reason="collaboration_flow_finished",
+        payload={"verification_status": state.get("verification_status")},
+    )
     state["messages"] = [*base_history, AIMessage(content=state["final_response"])]
     state["should_respond"] = True
     return state
@@ -2376,7 +2635,9 @@ def _can_delegate_within_hop_budget(state: AgentState) -> bool:
 
 def _stop_due_to_loop_guard(state: AgentState) -> AgentState:
     state["terminated_due_to_loop_guard"] = True
-    state["final_response"] = "这次需要处理的步骤有点多，我先停在这里。您可以把目标再明确一点，或让我先只完成其中一步。"
+    state["final_response"] = (
+        "这次需要处理的步骤有点多，我先停在这里。您可以把目标再明确一点，或让我先只完成其中一步。"
+    )
     state["messages"] = [*state.get("messages", []), AIMessage(content=state["final_response"])]
     return state
 
@@ -2386,7 +2647,9 @@ async def master_node(state: AgentState) -> AgentState:
     _set_phase(state, "phase_1_routing", reason="master_node_entered")
     _record_checkpoint(state, "routing.master_entered", reason="master_node_entered")
     user_messages = _filter_user_messages(state["messages"])
-    user_query = _stringify_message_content(user_messages[-1].content).strip() if user_messages else ""
+    user_query = (
+        _stringify_message_content(user_messages[-1].content).strip() if user_messages else ""
+    )
 
     state["current_agent"] = state.get("current_agent") or AgentRole.MASTER
     state["active_agents"] = _normalize_active_agents(state.get("active_agents"))
@@ -2425,7 +2688,9 @@ async def master_node(state: AgentState) -> AgentState:
     elif clarification_route is not None:
         state["execution_mode"] = "direct"
         routed_agent = clarification_route
-    elif _is_short_confirmation(user_query) and _previous_turn_proposed_schedule_creation(state.get("messages", [])):
+    elif _is_short_confirmation(user_query) and _previous_turn_proposed_schedule_creation(
+        state.get("messages", [])
+    ):
         state["execution_mode"] = "direct"
         routed_agent = AgentRole.SCHEDULE_PLANNER
     else:
@@ -2433,7 +2698,9 @@ async def master_node(state: AgentState) -> AgentState:
         state["routing_decision"] = routing_metadata
         if request_mode == "collaboration":
             collaboration_state = await _run_collaboration_flow(state, user_query)
-            if collaboration_state.get("execution_mode") == "collaboration" or collaboration_state.get("final_response"):
+            if collaboration_state.get(
+                "execution_mode"
+            ) == "collaboration" or collaboration_state.get("final_response"):
                 return collaboration_state
         state["execution_mode"] = "direct"
         routed_agent = _route_agent_from_user_query(user_query)
@@ -2445,11 +2712,16 @@ async def master_node(state: AgentState) -> AgentState:
         state["next_step"] = routed_agent.value
         if routed_agent not in state["active_agents"]:
             state["active_agents"] = [*state["active_agents"], routed_agent]
-        state["agent_trace"] = [*(state.get("agent_trace") or [AgentRole.MASTER.value]), routed_agent.value]
+        state["agent_trace"] = [
+            *(state.get("agent_trace") or [AgentRole.MASTER.value]),
+            routed_agent.value,
+        ]
         return state
 
     llm = _get_llm_for_state(state)
-    response = await _invoke_llm(llm, [SystemMessage(content=MASTER_SYSTEM_PROMPT), *state["messages"]])
+    response = await _invoke_llm(
+        llm, [SystemMessage(content=MASTER_SYSTEM_PROMPT), *state["messages"]]
+    )
     _record_response_usage(state, response)
     content = _stringify_message_content(response.content).strip()
 
@@ -2462,7 +2734,10 @@ async def master_node(state: AgentState) -> AgentState:
         state["next_step"] = routed_agent.value
         if routed_agent not in state["active_agents"]:
             state["active_agents"] = [*state["active_agents"], routed_agent]
-        state["agent_trace"] = [*(state.get("agent_trace") or [AgentRole.MASTER.value]), routed_agent.value]
+        state["agent_trace"] = [
+            *(state.get("agent_trace") or [AgentRole.MASTER.value]),
+            routed_agent.value,
+        ]
         return state
 
     state["final_response"] = content
@@ -2476,7 +2751,9 @@ async def planner_node(state: AgentState) -> AgentState:
     user_query = _stringify_message_content(user_messages[-1].content) if user_messages else ""
     if _has_active_clarification_context(state):
         user_query = _build_resumed_clarification_query(state, user_query)
-    elif _is_short_confirmation(user_query) and _previous_turn_proposed_schedule_creation(state.get("messages", [])):
+    elif _is_short_confirmation(user_query) and _previous_turn_proposed_schedule_creation(
+        state.get("messages", [])
+    ):
         user_query = _expand_schedule_confirmation_query(user_query, state.get("messages", []))
     return await _run_sub_commander(
         state,
@@ -2526,6 +2803,53 @@ async def analyst_node(state: AgentState) -> AgentState:
     )
 
 
+async def code_commander_node(state: AgentState) -> AgentState:
+    """
+    Code Commander node - handles code execution tasks
+    Routes to DirectExecutor (LOW risk) or SandboxExecutor (HIGH risk)
+    """
+    from app.agents.prompts import CODE_COMMANDER_SYSTEM_PROMPT
+    from app.agents.tools.ai_adapter import get_adapter
+    from app.agents.tools.direct_executor import DirectExecutor
+    from app.agents.tools.sandbox_executor import SandboxExecutor
+    from app.agents.tools.security_classifier import RiskLevel, SecurityClassifier
+
+    user_messages = _filter_user_messages(state["messages"])
+    user_query = _stringify_message_content(user_messages[-1].content) if user_messages else ""
+
+    # Get AI provider from state (default: claude)
+    ai_provider = str(state.get("code_ai_provider") or "claude")
+
+    # Classify risk level
+    classifier = SecurityClassifier()
+    risk_level = classifier.classify(user_query)
+
+    # Select executor based on risk
+    adapter = get_adapter(ai_provider)
+    if risk_level == RiskLevel.LOW:
+        executor = DirectExecutor(adapter)
+        result_parts = []
+        async for line in executor.execute(user_query):
+            result_parts.append(line)
+        result_output = "".join(result_parts)
+    else:
+        sandbox_exec = SandboxExecutor(adapter)
+        result_parts = []
+        async for line in sandbox_exec.execute(user_query):
+            result_parts.append(line)
+        result_output = "".join(result_parts)
+
+    state["final_response"] = result_output
+    state["code_execution_result"] = {
+        "ai_provider": ai_provider,
+        "risk_level": risk_level.value,
+        "output": result_output,
+    }
+    state["messages"] = [*state.get("messages", []), AIMessage(content=result_output)]
+    state["should_respond"] = True
+    return state
+
+
 def route_agent(state: AgentState) -> str:
     if state.get("final_response"):
         return END
@@ -2553,6 +2877,7 @@ def create_agent_graph(callbacks: list | None = None):
     graph.add_node(AgentRole.EXECUTOR.value, executor_node)
     graph.add_node(AgentRole.LIBRARIAN.value, librarian_node)
     graph.add_node(AgentRole.ANALYST.value, analyst_node)
+    graph.add_node(AgentRole.CODE_COMMANDER.value, code_commander_node)
 
     graph.set_entry_point(AgentRole.MASTER.value)
     graph.add_conditional_edges(
@@ -2563,6 +2888,7 @@ def create_agent_graph(callbacks: list | None = None):
             AgentRole.EXECUTOR.value: AgentRole.EXECUTOR.value,
             AgentRole.LIBRARIAN.value: AgentRole.LIBRARIAN.value,
             AgentRole.ANALYST.value: AgentRole.ANALYST.value,
+            AgentRole.CODE_COMMANDER.value: AgentRole.CODE_COMMANDER.value,
             END: END,
         },
     )
@@ -2572,6 +2898,7 @@ def create_agent_graph(callbacks: list | None = None):
         AgentRole.EXECUTOR,
         AgentRole.LIBRARIAN,
         AgentRole.ANALYST,
+        AgentRole.CODE_COMMANDER,
     ):
         graph.add_edge(role.value, END)
 
diff --git a/backend/app/agents/prompts.py b/backend/app/agents/prompts.py
index bcd6cd3..d1b2f06 100644
--- a/backend/app/agents/prompts.py
+++ b/backend/app/agents/prompts.py
@@ -309,14 +309,14 @@ ANALYST_INSIGHTS_PROMPT = f"""{JARVIS_PERSONA_PROMPT}
 
 你是 analyst 体系下的洞察建议官，负责从任务、论坛和知识线索里提炼趋势、风险与建议。
 
-## 允许使用的工具:
+## 你的允许使用的工具:
 - get_tasks
 - get_forum_posts
 - search_knowledge
 - hybrid_search
 - web_search
 
-## 要求:
+## 你的要求:
 - 先给结论与判断
 - 再说明依据与建议
 - 当需要外部/最新信息时，可使用 `web_search`
@@ -324,6 +324,38 @@ ANALYST_INSIGHTS_PROMPT = f"""{JARVIS_PERSONA_PROMPT}
 """
 
 
+CODE_COMMANDER_SYSTEM_PROMPT = f"""{JARVIS_PERSONA_PROMPT}
+
+你是代码指挥官，负责协调 AI 写代码助手。
+
+## 你的职责:
+1. 接收用户选择的 AI 提供商（Claude/Gemini/Codex/OpenCode）
+2. 接收用户的写代码需求
+3. 进行安全分级判定
+4. 路由到合适的执行器
+
+## 安全分级规则:
+- 低风险：demo、示例、贪食蛇游戏等独立项目
+- 高风险：修改现有项目、涉及 Jarvis 项目、路径操作等
+
+## 执行模式:
+- 直接执行：低风险任务，直接运行
+- 沙盒执行：高风险任务，在临时目录隔离执行
+
+## 你的输出:
+- 简洁汇报执行结果
+- 如果需要用户交互（如确认 "y"），明确提示
+"""
+
+
+SANDBOX_EXECUTION_PROMPT = """将在隔离的临时目录中执行任务。
+任务完成后，工作目录会被保留供下载。"""
+
+
+DIRECT_EXECUTION_PROMPT = """将直接执行任务。
+如果需要交互，请等待用户输入。"""
+
+
 COORDINATOR_SYSTEM_PROMPT = f"""{JARVIS_PERSONA_PROMPT}
 
 你是 Jarvis 的协作协调官，负责把复杂请求收束成最小受控协作，而不是放任系统进入自由 swarm。
@@ -382,6 +414,7 @@ TOP_LEVEL_SYSTEM_PROMPTS_BY_KEY = {
     "executor": EXECUTOR_SYSTEM_PROMPT,
     "librarian": LIBRARIAN_SYSTEM_PROMPT,
     "analyst": ANALYST_SYSTEM_PROMPT,
+    "code_commander": CODE_COMMANDER_SYSTEM_PROMPT,
 }
 
 
diff --git a/backend/app/agents/registry/builtins.py b/backend/app/agents/registry/builtins.py
index 7f9dd4c..a9b99d4 100644
--- a/backend/app/agents/registry/builtins.py
+++ b/backend/app/agents/registry/builtins.py
@@ -29,6 +29,7 @@ TOP_LEVEL_AGENT_DEFAULT_SUB_COMMANDERS: dict[str, tuple[str, ...]] = {
         "analyst_progress",
         "analyst_insights",
     ),
+    AgentRole.CODE_COMMANDER.value: (),
 }
 
 TOP_LEVEL_AGENT_DISPLAY_NAMES: dict[str, str] = {
@@ -37,6 +38,7 @@ TOP_LEVEL_AGENT_DISPLAY_NAMES: dict[str, str] = {
     AgentRole.EXECUTOR.value: "Executor",
     AgentRole.LIBRARIAN.value: "Librarian",
     AgentRole.ANALYST.value: "Analyst",
+    AgentRole.CODE_COMMANDER.value: "Code Commander",
 }
 
 TOP_LEVEL_AGENT_ROUTING_HINTS: dict[str, tuple[str, ...]] = {
@@ -55,6 +57,9 @@ TOP_LEVEL_AGENT_ROUTING_HINTS: dict[str, tuple[str, ...]] = {
     AgentRole.ANALYST.value: (
         "Handle reporting and insight requests using analyst sub-commanders.",
     ),
+    AgentRole.CODE_COMMANDER.value: (
+        "Handle code writing and execution tasks using AI CLI adapters.",
+    ),
 }
 
 TOP_LEVEL_AGENT_ALLOWED_SPAWN_ROLES: dict[str, tuple[str, ...]] = {
@@ -63,11 +68,13 @@ TOP_LEVEL_AGENT_ALLOWED_SPAWN_ROLES: dict[str, tuple[str, ...]] = {
         AgentRole.EXECUTOR.value,
         AgentRole.LIBRARIAN.value,
         AgentRole.ANALYST.value,
+        AgentRole.CODE_COMMANDER.value,
     ),
     AgentRole.SCHEDULE_PLANNER.value: (AgentRole.SCHEDULE_PLANNER.value,),
     AgentRole.EXECUTOR.value: (AgentRole.EXECUTOR.value,),
     AgentRole.LIBRARIAN.value: (AgentRole.LIBRARIAN.value,),
     AgentRole.ANALYST.value: (AgentRole.ANALYST.value,),
+    AgentRole.CODE_COMMANDER.value: (),
 }
 
 SUB_COMMANDER_PARENT_AGENT_IDS: dict[str, str] = {
@@ -99,11 +106,7 @@ BUILTIN_AGENT_MANIFESTS: tuple[AgentManifest, ...] = tuple(
 
 
 _capability_tool_names = tuple(
-    dict.fromkeys(
-        tool.name
-        for tools in SUB_COMMANDER_TOOLSETS.values()
-        for tool in tools
-    )
+    dict.fromkeys(tool.name for tools in SUB_COMMANDER_TOOLSETS.values() for tool in tools)
 )
 
 _CAPABILITY_METADATA_BY_TOOL_NAME: dict[str, dict[str, object]] = {
@@ -260,9 +263,7 @@ BUILTIN_SUB_COMMANDER_MANIFESTS: tuple[SubCommanderManifest, ...] = tuple(
         sub_commander_id=sub_commander_id,
         parent_agent_id=SUB_COMMANDER_PARENT_AGENT_IDS[sub_commander_id],
         prompt_text=SUB_COMMANDER_PROMPTS_BY_KEY[sub_commander_id],
-        capability_ids=list(
-            dict.fromkeys(tool.name for tool in tools)
-        ),
+        capability_ids=list(dict.fromkeys(tool.name for tool in tools)),
     )
     for sub_commander_id, tools in SUB_COMMANDER_TOOLSETS.items()
 )
diff --git a/backend/app/agents/schemas/task.py b/backend/app/agents/schemas/task.py
index dbea254..6e24132 100644
--- a/backend/app/agents/schemas/task.py
+++ b/backend/app/agents/schemas/task.py
@@ -1,7 +1,9 @@
 from __future__ import annotations
 
 from datetime import datetime, timezone
+from enum import Enum
 from typing import Any, Literal
+from uuid import uuid4
 
 from pydantic import BaseModel, Field
 
@@ -13,6 +15,18 @@ InterruptStatus = Literal["requested", "acknowledged", "resolved"]
 BudgetMode = Literal["direct", "collaboration"]
 
 
+class CodeProviderType(str, Enum):
+    CLAUDE = "claude"
+    GEMINI = "gemini"
+    CODEX = "codex"
+    OPENCODE = "opencode"
+
+
+class RiskLevelType(str, Enum):
+    LOW = "low"
+    HIGH = "high"
+
+
 class InterruptRecord(BaseModel):
     interrupt_id: str
     reason: str
@@ -83,3 +97,37 @@ class TaskResult(BaseModel):
     budget_snapshot: CollaborationBudget | dict[str, Any] | None = None
     next_action: str | None = None
     output_data: dict[str, Any] | None = None
+
+
+class CodeTaskType(str, Enum):
+    DEMO = "demo"
+    PROJECT = "project"
+    MODIFICATION = "modification"
+
+
+class CodeTask(BaseModel):
+    """代码任务请求模型"""
+
+    task_id: str = Field(default_factory=lambda: str(uuid4()))
+    task_type: CodeTaskType
+    ai_provider: CodeProviderType
+    sandbox_mode: bool = False
+    workspace_path: str | None = None
+    user_prompt: str
+    parent_task_id: str | None = None
+    thread_id: str | None = None
+    message_id: str | None = None
+    created_at: datetime = Field(default_factory=lambda: datetime.now(timezone.utc))
+
+
+class CodeExecutionResultSchema(BaseModel):
+    """代码执行结果模型 (API 响应用)"""
+
+    success: bool
+    message: str
+    files_created: list[str] = Field(default_factory=list)
+    output: str = ""
+    error: str | None = None
+    exit_code: int = 0
+    execution_time: float | None = None
+    sandbox_session_id: str | None = None
diff --git a/backend/app/agents/state.py b/backend/app/agents/state.py
index 773aad3..df58ca0 100644
--- a/backend/app/agents/state.py
+++ b/backend/app/agents/state.py
@@ -4,7 +4,14 @@ from typing import Annotated, Any, Literal, TypedDict
 
 from app.agents.schemas.event import AgentEvent
 from app.agents.schemas.message import AgentMessage
-from app.agents.schemas.task import AgentTask, CollaborationBudget, InterruptRecord, RecoveryRecord, TaskResult, VerificationStatus
+from app.agents.schemas.task import (
+    AgentTask,
+    CollaborationBudget,
+    InterruptRecord,
+    RecoveryRecord,
+    TaskResult,
+    VerificationStatus,
+)
 from langchain_core.messages import AIMessage, BaseMessage, HumanMessage
 from langgraph.graph.message import add_messages
 
@@ -23,6 +30,7 @@ class AgentRole(str, Enum):
     EXECUTOR = "executor"
     LIBRARIAN = "librarian"
     ANALYST = "analyst"
+    CODE_COMMANDER = "code_commander"
 
 
 @dataclass
@@ -141,6 +149,14 @@ class AgentState(TypedDict):
     user_llm_config: dict[str, Any] | None
     provider_capabilities: dict[str, Any] | None
 
+    # Code Commander state
+    code_task_type: Literal["demo", "project", "modification"] | None
+    code_ai_provider: Literal["claude", "gemini", "codex", "opencode"] | None
+    code_sandbox_mode: bool | None
+    code_workspace_path: str | None
+    code_execution_session_id: str | None
+    code_execution_result: dict[str, Any] | None
+
 
 def initial_state(user_id: str, conversation_id: str) -> AgentState:
     return AgentState(
diff --git a/backend/app/agents/tools/__init__.py b/backend/app/agents/tools/__init__.py
index eb114b6..68d1c0f 100644
--- a/backend/app/agents/tools/__init__.py
+++ b/backend/app/agents/tools/__init__.py
@@ -138,3 +138,12 @@ SUB_COMMANDER_TOOLSETS = {
     "analyst_progress": ANALYST_PROGRESS_TOOLS,
     "analyst_insights": ANALYST_INSIGHT_TOOLS,
 }
+
+# Code Commander toolset (tools implemented in later phases)
+CODE_COMMANDER_TOOLSET_NAMES = [
+    "execute_code_task",
+    "get_execution_status",
+    "send_interactive_input",
+    "download_workspace",
+    "cleanup_workspace",
+]
diff --git a/backend/app/agents/tools/ai_adapter.py b/backend/app/agents/tools/ai_adapter.py
new file mode 100644
index 0000000..06d3fbb
--- /dev/null
+++ b/backend/app/agents/tools/ai_adapter.py
@@ -0,0 +1,196 @@
+"""
+AI CLI Adapter - 统一接口适配不同 AI CLI (Claude/Gemini/Codex/OpenCode)
+"""
+
+from abc import ABC, abstractmethod
+from dataclasses import dataclass, field
+from pathlib import Path
+from typing import Literal
+
+
+@dataclass
+class CodeExecutionResult:
+    """代码执行结果"""
+
+    success: bool
+    message: str
+    files_created: list[str] = field(default_factory=list)
+    output: str = ""
+    error: str | None = None
+    exit_code: int = 0
+
+
+class AICLIAdapter(ABC):
+    """AI CLI 适配器抽象基类"""
+
+    @property
+    @abstractmethod
+    def cli_name(self) -> str:
+        """CLI 命令名称，如 'claude', 'gemini'"""
+        pass
+
+    @property
+    @abstractmethod
+    def requires_workspace(self) -> bool:
+        """是否需要工作目录"""
+        pass
+
+    @property
+    def provider(self) -> Literal["claude", "gemini", "codex", "opencode"]:
+        """AI 提供商标识"""
+        return self.cli_name
+
+    @abstractmethod
+    def build_command(self, prompt: str, workspace: Path | None) -> list[str]:
+        """构建 CLI 命令"""
+        pass
+
+    @abstractmethod
+    def parse_output(self, output: str) -> CodeExecutionResult:
+        """解析 CLI 输出"""
+        pass
+
+    @abstractmethod
+    def is_installed(self) -> bool:
+        """检查 CLI 是否已安装"""
+        pass
+
+
+class ClaudeAdapter(AICLIAdapter):
+    """Claude CLI 适配器"""
+
+    cli_name = "claude"
+    requires_workspace = True
+
+    def build_command(self, prompt: str, workspace: Path | None) -> list[str]:
+        cmd = ["claude", "-p", prompt]
+        if workspace:
+            cmd.extend(["--output-format", "stream-json"])
+        cmd.append("--dangerously-skip-permissions")
+        return cmd
+
+    def parse_output(self, output: str) -> CodeExecutionResult:
+        # Claude CLI 输出可能是纯文本或 JSON
+        # 简化处理：直接返回输出
+        if not output.strip():
+            return CodeExecutionResult(
+                success=False,
+                message="No output from Claude CLI",
+                output=output,
+            )
+        return CodeExecutionResult(
+            success=True,
+            message="Execution completed",
+            output=output,
+        )
+
+    def is_installed(self) -> bool:
+        import shutil
+
+        return shutil.which("claude") is not None
+
+
+class GeminiAdapter(AICLIAdapter):
+    """Gemini CLI 适配器"""
+
+    cli_name = "gemini"
+    requires_workspace = False
+
+    def build_command(self, prompt: str, workspace: Path | None) -> list[str]:
+        cmd = ["gemini", "-p", prompt]
+        return cmd
+
+    def parse_output(self, output: str) -> CodeExecutionResult:
+        if not output.strip():
+            return CodeExecutionResult(
+                success=False,
+                message="No output from Gemini CLI",
+                output=output,
+            )
+        return CodeExecutionResult(
+            success=True,
+            message="Execution completed",
+            output=output,
+        )
+
+    def is_installed(self) -> bool:
+        import shutil
+
+        return shutil.which("gemini") is not None
+
+
+class CodexAdapter(AICLIAdapter):
+    """Codex CLI 适配器"""
+
+    cli_name = "codex"
+    requires_workspace = True
+
+    def build_command(self, prompt: str, workspace: Path | None) -> list[str]:
+        cmd = ["codex", "-p", prompt]
+        return cmd
+
+    def parse_output(self, output: str) -> CodeExecutionResult:
+        if not output.strip():
+            return CodeExecutionResult(
+                success=False,
+                message="No output from Codex CLI",
+                output=output,
+            )
+        return CodeExecutionResult(
+            success=True,
+            message="Execution completed",
+            output=output,
+        )
+
+    def is_installed(self) -> bool:
+        import shutil
+
+        return shutil.which("codex") is not None
+
+
+class OpenCodeAdapter(AICLIAdapter):
+    """OpenCode CLI 适配器"""
+
+    cli_name = "opencode"
+    requires_workspace = True
+
+    def build_command(self, prompt: str, workspace: Path | None) -> list[str]:
+        cmd = ["opencode", "-p", prompt]
+        return cmd
+
+    def parse_output(self, output: str) -> CodeExecutionResult:
+        if not output.strip():
+            return CodeExecutionResult(
+                success=False,
+                message="No output from OpenCode CLI",
+                output=output,
+            )
+        return CodeExecutionResult(
+            success=True,
+            message="Execution completed",
+            output=output,
+        )
+
+    def is_installed(self) -> bool:
+        import shutil
+
+        return shutil.which("opencode") is not None
+
+
+# 提供商注册表
+ADAPTER_REGISTRY: dict[str, AICLIAdapter] = {
+    "claude": ClaudeAdapter(),
+    "gemini": GeminiAdapter(),
+    "codex": CodexAdapter(),
+    "opencode": OpenCodeAdapter(),
+}
+
+
+def get_adapter(provider: str) -> AICLIAdapter:
+    """获取指定提供商的适配器"""
+    adapter = ADAPTER_REGISTRY.get(provider.lower())
+    if adapter is None:
+        raise ValueError(
+            f"Unknown AI provider: {provider}. Available: {list(ADAPTER_REGISTRY.keys())}"
+        )
+    return adapter
diff --git a/backend/app/agents/tools/direct_executor.py b/backend/app/agents/tools/direct_executor.py
new file mode 100644
index 0000000..79ac6cb
--- /dev/null
+++ b/backend/app/agents/tools/direct_executor.py
@@ -0,0 +1,112 @@
+"""
+Direct Executor - 直接执行器
+用于低风险任务，直接执行不隔离
+"""
+
+import asyncio
+import os
+import shutil
+import tempfile
+from pathlib import Path
+from typing import AsyncGenerator
+
+from app.agents.tools.ai_adapter import AICLIAdapter
+
+
+class ExecutionResult:
+    """执行结果"""
+
+    def __init__(
+        self,
+        success: bool,
+        exit_code: int,
+        stdout: str,
+        stderr: str,
+    ):
+        self.success = success
+        self.exit_code = exit_code
+        self.stdout = stdout
+        self.stderr = stderr
+
+
+class DirectExecutor:
+    """直接执行器（用于低风险任务）"""
+
+    def __init__(self, adapter: AICLIAdapter, timeout: int = 60):
+        self.adapter = adapter
+        self.timeout = timeout
+
+    async def execute(
+        self,
+        prompt: str,
+    ) -> AsyncGenerator[str, None]:
+        """
+        直接执行，不需要沙盒
+
+        Args:
+            prompt: 任务描述
+
+        Yields:
+            str: 实时输出
+        """
+        # 1. 检查 CLI 是否安装
+        if not self.adapter.is_installed():
+            yield f"[ERROR] {self.adapter.cli_name} is not installed\n"
+            yield f"[ERROR] Please install {self.adapter.cli_name} first\n"
+            return
+
+        # 2. 构建命令
+        cmd = self.adapter.build_command(prompt, None)
+
+        # 3. 异步执行，实时 yield 输出
+        process = await asyncio.create_subprocess_exec(
+            *cmd,
+            stdout=asyncio.subprocess.PIPE,
+            stderr=asyncio.subprocess.PIPE,
+            env={**os.environ, "TERM": "xterm-256color"},
+        )
+
+        # 4. 实时读取输出
+        stdout_lines = []
+        stderr_lines = []
+
+        while True:
+            try:
+                line_bytes = await asyncio.wait_for(
+                    process.stdout.readline(),
+                    timeout=self.timeout,
+                )
+                if not line_bytes:
+                    break
+                line = line_bytes.decode("utf-8", errors="replace")
+                stdout_lines.append(line)
+                yield line
+            except asyncio.TimeoutError:
+                process.kill()
+                yield f"\n[ERROR] Execution timed out after {self.timeout}s\n"
+                break
+
+        # 5. 读取 stderr
+        stderr_bytes = await process.communicate()
+        if stderr_bytes[1]:
+            stderr = stderr_bytes[1].decode("utf-8", errors="replace")
+            stderr_lines.append(stderr)
+            yield f"\n[STDERR]\n{stderr}\n"
+
+        # 6. 完成标记
+        yield f"\n[EXIT_CODE] {process.returncode or 0}\n"
+        yield f"\n[COMPLETE] success={process.returncode == 0}\n"
+
+    async def execute_sync(self, prompt: str) -> ExecutionResult:
+        """同步执行并返回完整结果"""
+        output_parts = []
+        async for line in self.execute(prompt):
+            output_parts.append(line)
+
+        output = "".join(output_parts)
+        return ExecutionResult(
+            success="[COMPLETE] success=True" in output,
+            exit_code=0,
+            stdout=output,
+            stderr="",
+        )
diff --git a/backend/app/agents/tools/interactive_input.py b/backend/app/agents/tools/interactive_input.py
new file mode 100644
index 0000000..3b7ff4f
--- /dev/null
+++ b/backend/app/agents/tools/interactive_input.py
@@ -0,0 +1,58 @@
+"""
+InteractiveInputHandler - 交互输入处理
+"""
+
+import asyncio
+from typing import Any
+
+from app.agents.tools.terminal_engine import PTYManager
+
+
+class InteractiveInputHandler:
+    """交互输入处理器"""
+
+    def __init__(self, pty_manager: PTYManager):
+        self.pty_manager = pty_manager
+        self._pending_inputs: dict[str, asyncio.Event] = {}
+        self._input_cache: dict[str, str] = {}
+
+    async def wait_for_input(self, session_id: str, prompt: str) -> str:
+        """等待用户输入（如 "y" 确认）"""
+        event = asyncio.Event()
+        self._pending_inputs[session_id] = event
+
+        # 发送提示
+        from app.routers.terminal import manager
+
+        try:
+            await manager.send(session_id, f"\n{prompt}\n")
+        except Exception:
+            pass
+
+        # 等待输入完成
+        try:
+            await asyncio.wait_for(event.wait(), timeout=60.0)
+        except asyncio.TimeoutError:
+            del self._pending_inputs[session_id]
+            return self._input_cache.get(session_id, "")
+
+        del self._pending_inputs[session_id]
+
+        return self._input_cache.get(session_id, "")
+
+    async def send_input(self, session_id: str, data: str):
+        """用户发送输入"""
+        self._input_cache[session_id] = data
+        if session_id in self._pending_inputs:
+            self._pending_inputs[session_id].set()
+
+        # 同时写入 PTY
+        await self.pty_manager.write(session_id, data + "\n")
+
+    def clear_input(self, session_id: str):
+        """清除输入缓存"""
+        if session_id in self._input_cache:
+            del self._input_cache[session_id]
+        if session_id in self._pending_inputs:
+            self._pending_inputs[session_id].set()  # 取消等待
+            del self._pending_inputs[session_id]
diff --git a/backend/app/agents/tools/sandbox_executor.py b/backend/app/agents/tools/sandbox_executor.py
new file mode 100644
index 0000000..ed3b1b4
--- /dev/null
+++ b/backend/app/agents/tools/sandbox_executor.py
@@ -0,0 +1,173 @@
+"""
+Sandbox Executor - 沙盒执行器
+在高风险任务在隔离的临时目录中执行
+"""
+
+import asyncio
+import os
+import shutil
+import tempfile
+from dataclasses import dataclass, field
+from pathlib import Path
+from typing import AsyncGenerator
+
+from app.agents.tools.ai_adapter import AICLIAdapter, CodeExecutionResult
+
+
+@dataclass
+class SandboxEnvironment:
+    """沙盒环境"""
+
+    workspace_path: Path
+    session_id: str
+
+    @staticmethod
+    async def create(prefix: str = "jarvis_code_") -> "SandboxEnvironment":
+        """创建新的沙盒环境"""
+        temp_dir = tempfile.mkdtemp(prefix=prefix)
+        session_id = Path(temp_dir).name
+        return SandboxEnvironment(
+            workspace_path=Path(temp_dir),
+            session_id=session_id,
+        )
+
+    async def cleanup(self) -> None:
+        """清理沙盒环境"""
+        if self.workspace_path.exists():
+            shutil.rmtree(self.workspace_path)
+
+    def list_created_files(self) -> list[str]:
+        """列出沙盒中创建的文件"""
+        if not self.workspace_path.exists():
+            return []
+        files = []
+        for root, dirs, filenames in os.walk(self.workspace_path):
+            for filename in filenames:
+                full_path = os.path.join(root, filename)
+                rel_path = os.path.relpath(full_path, self.workspace_path)
+                files.append(rel_path)
+        return files
+
+
+@dataclass
+class ExecutionResult:
+    """执行结果"""
+
+    success: bool
+    exit_code: int
+    stdout: str
+    stderr: str
+    files_created: list[str] = field(default_factory=list)
+
+
+class SandboxExecutor:
+    """沙盒执行器"""
+
+    def __init__(self, adapter: AICLIAdapter, timeout: int = 300):
+        self.adapter = adapter
+        self.timeout = timeout
+        self._sessions: dict[str, SandboxEnvironment] = {}
+
+    async def execute(
+        self,
+        prompt: str,
+        session_id: str | None = None,
+    ) -> AsyncGenerator[str, None]:
+        """
+        执行代码任务，yield 实时输出
+
+        Args:
+            prompt: 任务描述
+            session_id: 会话 ID（可选，用于复用沙盒）
+
+        Yields:
+            str: 实时输出
+        """
+        # 1. 创建或复用沙盒环境
+        if session_id and session_id in self._sessions:
+            env = self._sessions[session_id]
+        else:
+            env = await SandboxEnvironment.create()
+            self._sessions[env.session_id] = env
+            session_id = env.session_id
+
+        # 2. 构建命令
+        cmd = self.adapter.build_command(prompt, env.workspace_path)
+
+        # 3. 异步执行，实时 yield 输出
+        process = await asyncio.create_subprocess_exec(
+            *cmd,
+            stdout=asyncio.subprocess.PIPE,
+            stderr=asyncio.subprocess.PIPE,
+            cwd=str(env.workspace_path),
+            env={**os.environ, "TERM": "xterm-256color"},
+        )
+
+        # 4. 实时读取输出
+        stdout_lines = []
+        stderr_lines = []
+
+        while True:
+            try:
+                line_bytes = await asyncio.wait_for(
+                    process.stdout.readline(),
+                    timeout=self.timeout,
+                )
+                if not line_bytes:
+                    break
+                line = line_bytes.decode("utf-8", errors="replace")
+                stdout_lines.append(line)
+                yield line
+            except asyncio.TimeoutError:
+                process.kill()
+                yield f"\n[ERROR] Execution timed out after {self.timeout}s\n"
+                break
+
+        # 5. 读取 stderr
+        stderr_bytes = await process.communicate()
+        if stderr_bytes[1]:
+            stderr = stderr_bytes[1].decode("utf-8", errors="replace")
+            stderr_lines.append(stderr)
+            yield f"\n[STDERR]\n{stderr}\n"
+
+        # 6. 返回结果（通过 yield 完成标记）
+        result = ExecutionResult(
+            success=process.returncode == 0,
+            exit_code=process.returncode or 0,
+            stdout="".join(stdout_lines),
+            stderr="".join(stderr_lines),
+            files_created=env.list_created_files(),
+        )
+        yield f"\n[EXIT_CODE] {result.exit_code}\n"
+        yield f"\n[COMPLETE] success={result.success}\n"
+
+    async def get_result(
+        self,
+        prompt: str,
+        session_id: str | None = None,
+    ) -> ExecutionResult:
+        """同步执行并返回完整结果"""
+        output_parts = []
+        async for line in self.execute(prompt, session_id):
+            output_parts.append(line)
+        # 解析最后的结果
+        # 实际使用中可能需要更复杂的结果收集
+        return ExecutionResult(
+            success="[COMPLETE] success=True" in "".join(output_parts),
+            exit_code=0,
+            stdout="".join(output_parts),
+            stderr="",
+            files_created=[],
+        )
+
+    async def cleanup_session(self, session_id: str) -> bool:
+        """清理指定会话"""
+        if session_id in self._sessions:
+            await self._sessions[session_id].cleanup()
+            del self._sessions[session_id]
+            return True
+        return False
+
+    def get_session(self, session_id: str) -> SandboxEnvironment | None:
+        """获取会话环境"""
+        return self._sessions.get(session_id)
diff --git a/backend/app/agents/tools/security_classifier.py b/backend/app/agents/tools/security_classifier.py
new file mode 100644
index 0000000..b4ddc3b
--- /dev/null
+++ b/backend/app/agents/tools/security_classifier.py
@@ -0,0 +1,129 @@
+"""
+Security Classifier - 安全分级判定
+低风险任务直接执行，高风险任务沙盒执行
+"""
+
+from enum import Enum
+
+
+class RiskLevel(Enum):
+    LOW = "low"  # 直接执行
+    HIGH = "high"  # 沙盒执行
+
+
+class SecurityClassifier:
+    """安全分级器"""
+
+    HIGH_RISK_KEYWORDS = [
+        # 路径/项目操作
+        "修改",
+        "编辑",
+        "删除",
+        "移动",
+        "重命名",
+        "Jarvis",
+        "backend",
+        "frontend",
+        "git",
+        "config",
+        ".env",
+        "生产环境",
+        # 文件操作
+        "写入",
+        "创建文件在",
+        "移动到",
+        "提交",
+        "push",
+        "pull",
+        "merge",
+        # 系统操作
+        "sudo",
+        "rm ",
+        "chmod",
+        "chown",
+    ]
+
+    LOW_RISK_KEYWORDS = [
+        # demo/示例类
+        "demo",
+        "示例",
+        "贪食蛇",
+        "俄罗斯方块",
+        "小游戏",
+        "独立项目",
+        "新项目",
+        "创建一个",
+        "写一个",
+        "帮我写一个",
+        # 明确无害的请求
+        "生成代码",
+        "代码示例",
+        "练习项目",
+    ]
+
+    def classify(self, task_description: str, target_path: str | None = None) -> RiskLevel:
+        """
+        判断任务风险等级
+
+        Args:
+            task_description: 任务描述
+            target_path: 目标路径（如果有）
+
+        Returns:
+            RiskLevel: LOW 或 HIGH
+        """
+        # 1. 检查高风险关键词
+        task_lower = task_description.lower()
+        if any(kw.lower() in task_lower for kw in self.HIGH_RISK_KEYWORDS):
+            return RiskLevel.HIGH
+
+        # 2. 检查目标路径
+        if target_path and self._is_project_path(target_path):
+            return RiskLevel.HIGH
+
+        # 3. 检查低风险关键词
+        if any(kw.lower() in task_lower for kw in self.LOW_RISK_KEYWORDS):
+            return RiskLevel.LOW
+
+        # 4. 默认高风险（保守策略）
+        return RiskLevel.HIGH
+
+    def _is_project_path(self, path: str) -> bool:
+        """
+        检查路径是否指向项目目录
+
+        Args:
+            path: 文件路径
+
+        Returns:
+            bool: 如果是项目路径返回 True
+        """
+        path_lower = path.lower()
+        project_indicators = [
+            "jarvis",
+            "backend/app",
+            "frontend/src",
+            ".git",
+            "package.json",
+            "pyproject.toml",
+            "requirements.txt",
+        ]
+        return any(indicator in path_lower for indicator in project_indicators)
+
+    def get_risk_factors(
+        self, task_description: str, target_path: str | None = None
+    ) -> dict[str, bool]:
+        """
+        获取详细的风险因素分析
+
+        Returns:
+            dict: 各风险因素及其状态
+        """
+        task_lower = task_description.lower()
+        return {
+            "has_high_risk_keywords": any(
+                kw.lower() in task_lower for kw in self.HIGH_RISK_KEYWORDS
+            ),
+            "has_low_risk_keywords": any(kw.lower() in task_lower for kw in self.LOW_RISK_KEYWORDS),
+            "is_project_path": bool(target_path and self._is_project_path(target_path)),
+        }
diff --git a/backend/app/agents/tools/stream_output.py b/backend/app/agents/tools/stream_output.py
new file mode 100644
index 0000000..54bf19f
--- /dev/null
+++ b/backend/app/agents/tools/stream_output.py
@@ -0,0 +1,86 @@
+"""
+StreamOutput - 流式输出封装
+"""
+
+import json
+from datetime import datetime, timezone
+from typing import Any, AsyncGenerator, Callable
+
+from dataclasses import dataclass
+
+
+@dataclass
+class StreamEvent:
+    """流式事件"""
+
+    type: str  # "output" | "error" | "status" | "complete"
+    session_id: str
+    data: str
+    timestamp: str
+
+
+class StreamOutput:
+    """流式输出处理器"""
+
+    def __init__(
+        self,
+        session_id: str,
+        websocket_sender: Callable[[str, str], Any] | None = None,
+    ):
+        self.session_id = session_id
+        self.websocket_sender = websocket_sender
+        self._listeners: list[Callable[[StreamEvent], Any]] = []
+
+    def add_listener(self, listener: Callable[[StreamEvent], Any]):
+        """添加事件监听器"""
+        self._listeners.append(listener)
+
+    def remove_listener(self, listener: Callable[[StreamEvent], Any]):
+        """移除事件监听器"""
+        if listener in self._listeners:
+            self._listeners.remove(listener)
+
+    async def push(self, event_type: str, data: str):
+        """推送事件到 WebSocket 和监听器"""
+        event = StreamEvent(
+            type=event_type,
+            session_id=self.session_id,
+            data=data,
+            timestamp=datetime.now(timezone.utc).isoformat(),
+        )
+
+        # 发送到 WebSocket
+        if self.websocket_sender:
+            try:
+                await self.websocket_sender(self.session_id, json.dumps(event.__dict__))
+            except Exception:
+                pass
+
+        # 通知监听器
+        for listener in self._listeners:
+            try:
+                result = listener(event)
+                if hasattr(result, "__await__"):
+                    await result
+            except Exception:
+                pass
+
+    async def stream_execution(
+        self,
+        executor,
+        prompt: str,
+    ) -> AsyncGenerator[str, None]:
+        """包装执行器，实现流式输出"""
+        async for line in executor.execute(prompt):
+            await self.push("output", line)
+            yield line
+
+        await self.push("complete", "")
+
+    async def push_status(self, status: str):
+        """推送状态消息"""
+        await self.push("status", status)
+
+    async def push_error(self, error: str):
+        """推送错误消息"""
+        await self.push("error", error)
diff --git a/backend/app/agents/tools/terminal_engine.py b/backend/app/agents/tools/terminal_engine.py
new file mode 100644
index 0000000..9dd2f63
--- /dev/null
+++ b/backend/app/agents/tools/terminal_engine.py
@@ -0,0 +1,160 @@
+"""
+PTY Terminal Engine - 跨平台 PTY 终端管理
+"""
+
+import asyncio
+import os
+from dataclasses import dataclass, field
+from typing import AsyncGenerator
+
+from uuid import uuid4
+
+
+@dataclass
+class PTYSession:
+    """PTY 会话"""
+
+    session_id: str
+    process: asyncio.subprocess.Process
+    workspace_path: str
+
+
+class PTYManager:
+    """PTY 会话管理器"""
+
+    def __init__(self):
+        self._sessions: dict[str, PTYSession] = {}
+        self._output_queues: dict[str, asyncio.Queue] = {}
+
+    async def spawn(
+        self,
+        cli: str,
+        args: list[str],
+        cwd: str,
+        session_id: str | None = None,
+        env: dict | None = None,
+    ) -> str:
+        """启动 PTY 会话"""
+        if session_id is None:
+            session_id = f"pty_{uuid4().hex[:8]}"
+
+        # 构建环境变量
+        process_env = {**os.environ, "TERM": "xterm-256color"}
+        if env:
+            process_env.update(env)
+
+        # 创建 PTY 进程
+        process = await asyncio.create_subprocess_exec(
+            cli,
+            *args,
+            stdout=asyncio.subprocess.PIPE,
+            stderr=asyncio.subprocess.PIPE,
+            cwd=cwd,
+            env=process_env,
+        )
+
+        session = PTYSession(
+            session_id=session_id,
+            process=process,
+            workspace_path=cwd,
+        )
+        self._sessions[session_id] = session
+        self._output_queues[session_id] = asyncio.Queue()
+
+        # 启动输出读取协程
+        asyncio.create_task(self._read_output(session_id))
+
+        return session_id
+
+    async def _read_output(self, session_id: str):
+        """读取 PTY 输出并放入队列"""
+        session = self._sessions.get(session_id)
+        if not session:
+            return
+
+        queue = self._output_queues[session_id]
+
+        try:
+            while True:
+                line = await session.process.stdout.readline()
+                if not line:
+                    break
+                decoded_line = line.decode(errors="replace")
+                await queue.put(decoded_line)
+
+                # 广播到 WebSocket
+                await self._broadcast(session_id, decoded_line)
+
+            # 读取 stderr
+            stderr_line = await session.process.stderr.readline()
+            if stderr_line:
+                decoded_err = stderr_line.decode(errors="replace")
+                await queue.put(decoded_err)
+                await self._broadcast(session_id, decoded_err)
+
+        except Exception:
+            pass
+        finally:
+            await queue.put(None)  # 结束标记
+
+    async def write(self, session_id: str, data: str):
+        """写入 PTY（用户输入）"""
+        session = self._sessions.get(session_id)
+        if session and session.process.stdin:
+            session.process.stdin.write(data)
+            await session.process.stdin.drain()
+
+    async def read(self, session_id: str) -> AsyncGenerator[str, None]:
+        """读取 PTY 输出"""
+        queue = self._output_queues.get(session_id)
+        if not queue:
+            return
+
+        while True:
+            line = await queue.get()
+            if line is None:
+                break
+            yield line
+
+    async def resize(self, session_id: str, rows: int, cols: int):
+        """调整终端大小"""
+        # TODO: 实现 resize (需要平台特定实现)
+        pass
+
+    async def kill(self, session_id: str):
+        """终止 PTY 会话"""
+        if session_id in self._sessions:
+            session = self._sessions[session_id]
+            try:
+                session.process.terminate()
+                await asyncio.wait_for(session.process.wait(), timeout=3.0)
+            except asyncio.TimeoutError:
+                session.process.kill()
+                await session.process.wait()
+            except Exception:
+                pass
+            finally:
+                del self._sessions[session_id]
+                if session_id in self._output_queues:
+                    del self._output_queues[session_id]
+
+    async def _broadcast(self, session_id: str, data: str):
+        """广播输出到 WebSocket"""
+        from app.routers.terminal import manager
+
+        try:
+            await manager.send(session_id, data)
+        except Exception:
+            pass
+
+    def get_session(self, session_id: str) -> PTYSession | None:
+        """获取会话"""
+        return self._sessions.get(session_id)
+
+    def list_sessions(self) -> list[str]:
+        """列出所有会话 ID"""
+        return list(self._sessions.keys())
+
+
+# 全局单例
+pty_manager = PTYManager()
diff --git a/backend/app/main.py b/backend/app/main.py
index e695841..b5922d2 100644
--- a/backend/app/main.py
+++ b/backend/app/main.py
@@ -28,6 +28,7 @@ from app.routers import (
     marketplace_router,
     agent_skills_router,
     agent_sessions_router,
+    terminal_router,
 )
 from app.routers.scheduler import router as scheduler_router
 from app.services.scheduler_service import start_scheduler, stop_scheduler, get_scheduler_status
@@ -127,6 +128,7 @@ app.include_router(plugins_router)
 app.include_router(marketplace_router)
 app.include_router(agent_skills_router)
 app.include_router(agent_sessions_router)
+app.include_router(terminal_router)
 
 
 @app.get("/api/health")
diff --git a/backend/app/routers/__init__.py b/backend/app/routers/__init__.py
index c5dbad3..1c70869 100644
--- a/backend/app/routers/__init__.py
+++ b/backend/app/routers/__init__.py
@@ -20,3 +20,4 @@ from app.routers.plugins import router as plugins_router
 from app.routers.plugins import _marketplace_router as marketplace_router
 from app.routers.agent_skills import router as agent_skills_router
 from app.routers.agent_sessions import router as agent_sessions_router
+from app.routers.terminal import router as terminal_router
diff --git a/backend/app/routers/terminal.py b/backend/app/routers/terminal.py
new file mode 100644
index 0000000..8a435f2
--- /dev/null
+++ b/backend/app/routers/terminal.py
@@ -0,0 +1,79 @@
+"""
+Terminal WebSocket Router - 终端 WebSocket 端点
+"""
+
+from fastapi import APIRouter, WebSocket, WebSocketDisconnect
+
+from app.agents.tools.terminal_engine import pty_manager
+
+router = APIRouter(prefix="/ws/terminal", tags=["terminal"])
+
+
+class ConnectionManager:
+    """WebSocket 连接管理器"""
+
+    def __init__(self):
+        self.active_connections: dict[str, WebSocket] = {}
+
+    async def connect(self, session_id: str, websocket: WebSocket):
+        await websocket.accept()
+        self.active_connections[session_id] = websocket
+
+    def disconnect(self, session_id: str):
+        if session_id in self.active_connections:
+            del self.active_connections[session_id]
+
+    async def send(self, session_id: str, data: str):
+        if session_id in self.active_connections:
+            await self.active_connections[session_id].send_text(data)
+
+    def is_connected(self, session_id: str) -> bool:
+        return session_id in self.active_connections
+
+
+manager = ConnectionManager()
+
+
+@router.websocket("/{session_id}")
+async def terminal_websocket(websocket: WebSocket, session_id: str):
+    """终端 WebSocket 端点"""
+    await manager.connect(session_id, websocket)
+
+    try:
+        # 获取该 session 的输出队列
+        queue = pty_manager._output_queues.get(session_id)
+        if queue:
+            # 异步任务：转发 PTY 输出到 WebSocket
+            async def forward_output():
+                while True:
+                    try:
+                        data = await asyncio.wait_for(queue.get(), timeout=0.1)
+                        if data is None:
+                            await manager.send(session_id, "[SESSION_END]")
+                            break
+                        await manager.send(session_id, data)
+                    except asyncio.TimeoutError:
+                        # 检查连接是否还活跃
+                        if not manager.is_connected(session_id):
+                            break
+                    except Exception:
+                        break
+
+            import asyncio
+
+            forward_task = asyncio.create_task(forward_output())
+
+            # 主循环：接收用户输入
+            try:
+                while True:
+                    data = await websocket.receive_text()
+                    # 接收用户输入，写入 PTY
+                    await pty_manager.write(session_id, data + "\n")
+            except WebSocketDisconnect:
+                pass
+            finally:
+                forward_task.cancel()
+    except Exception:
+        pass
+    finally:
+        manager.disconnect(session_id)
diff --git a/development-doc/plan/code-update/checklist.md b/development-doc/plan/code-update/checklist.md
index d882525..338fe3e 100644
--- a/development-doc/plan/code-update/checklist.md
+++ b/development-doc/plan/code-update/checklist.md
@@ -22,13 +22,13 @@
 
 Day 1 目标：完成代码指挥官 Agent 的基础架子
 
-- [ ] 新增 `CODE_COMMANDER = "code_commander"` 到 `AgentRole` 枚举
-- [ ] 新增 `CodeCommanderState` TypedDict（包含 task_type, ai_provider, sandbox_mode 等）
-- [ ] 新增 `CODE_COMMANDER_SYSTEM_PROMPT` 系统提示
-- [ ] 新增 `SANDBOX_EXECUTION_PROMPT` 沙盒执行说明
-- [ ] 新增 `DIRECT_EXECUTION_PROMPT` 直接执行说明
-- [ ] 在 `SUB_COMMANDER_TOOLSETS` 中注册 `CODE_COMMANDER_TOOLSET`
-- [ ] 新增 `CodeCommanderManifest` 到 `AGENT_MANIFESTS`
+- [x] 新增 `CODE_COMMANDER = "code_commander"` 到 `AgentRole` 枚举
+- [x] 新增 `CodeCommanderState` TypedDict（包含 task_type, ai_provider, sandbox_mode 等）
+- [x] 新增 `CODE_COMMANDER_SYSTEM_PROMPT` 系统提示
+- [x] 新增 `SANDBOX_EXECUTION_PROMPT` 沙盒执行说明
+- [x] 新增 `DIRECT_EXECUTION_PROMPT` 直接执行说明
+- [x] 在 `SUB_COMMANDER_TOOLSETS` 中注册 `CODE_COMMANDER_TOOLSET`
+- [x] 新增 `CodeCommanderManifest` 到 `AGENT_MANIFESTS`
 - [ ] 补 Phase 1 单元测试
 
 **验收：确认 `AgentRole.CODE_COMMANDER` 存在且值正确**
@@ -39,17 +39,17 @@ Day 1 目标：完成代码指挥官 Agent 的基础架子
 
 Day 2 目标：实现适配不同 AI CLI 的统一接口
 
-- [ ] 新增 `AICLIAdapter` 抽象基类
+- [x] 新增 `AICLIAdapter` 抽象基类
   - `cli_name` 属性
   - `requires_workspace` 属性
   - `build_command()` 方法
   - `parse_output()` 方法
   - `is_installed()` 方法
-- [ ] 新增 `ClaudeAdapter` 实现
-- [ ] 新增 `GeminiAdapter` 实现
-- [ ] 新增 `CodexAdapter` 实现
-- [ ] 新增 `OpenCodeAdapter` 实现
-- [ ] 新增 `CodeExecutionResult` 数据类
+- [x] 新增 `ClaudeAdapter` 实现
+- [x] 新增 `GeminiAdapter` 实现
+- [x] 新增 `CodexAdapter` 实现
+- [x] 新增 `OpenCodeAdapter` 实现
+- [x] 新增 `CodeExecutionResult` 数据类
 - [ ] 补 Day 2 单元测试
 
 **验收：`AICLIAdapter` 可以正确识别 4 种 CLI**
@@ -60,13 +60,13 @@ Day 2 目标：实现适配不同 AI CLI 的统一接口
 
 Day 3 目标：实现安全分级和直接执行器
 
-- [ ] 新增 `RiskLevel` 枚举（LOW/HIGH）
-- [ ] 新增 `SecurityClassifier` 类
+- [x] 新增 `RiskLevel` 枚举（LOW/HIGH）
+- [x] 新增 `SecurityClassifier` 类
   - `HIGH_RISK_KEYWORDS` 列表
   - `LOW_RISK_KEYWORDS` 列表
   - `classify()` 方法实现
   - `_is_project_path()` 方法实现
-- [ ] 新增 `DirectExecutor` 类
+- [x] 新增 `DirectExecutor` 类
   - `execute()` 方法（异步）
   - 超时控制
   - `is_installed()` 检查
@@ -80,16 +80,16 @@ Day 3 目标：实现安全分级和直接执行器
 
 Day 4 目标：实现沙盒执行器
 
-- [ ] 新增 `SandboxEnvironment` 类
+- [x] 新增 `SandboxEnvironment` 类
   - `create()` 静态方法（创建临时目录）
   - `cleanup()` 方法
   - `workspace_path` 属性
   - `session_id` 属性
-- [ ] 新增 `SandboxExecutor` 类
+- [x] 新增 `SandboxExecutor` 类
   - `execute()` 方法（异步，yield 流式输出）
   - `cleanup_session()` 方法
   - `_list_created_files()` 方法
-- [ ] 实现超时控制
+- [x] 实现超时控制
 - [ ] 补 Day 4 单元测试
 
 **验收：`SandboxExecutor` 能创建、执行、清理沙盒**
@@ -115,15 +115,15 @@ Day 5 目标：确保执行引擎各组件协同工作
 
 Day 6 目标：将代码指挥官接入 LangGraph
 
-- [ ] 新增 `code_commander_node` 函数
+- [x] 新增 `code_commander_node` 函数
   - 获取用户需求和 AI 提供商
   - 调用 `SecurityClassifier`
   - 根据风险等级选择执行器
   - 返回执行结果
-- [ ] 在 `NODES` 字典中注册 `code_commander`
-- [ ] 新增 `_should_route_to_code_commander()` 路由函数
-- [ ] 在 `graph.py` 中添加条件边
-- [ ] 新增 `CodeTask`, `CodeExecutionResult` 模型到 `schemas/task.py`
+- [x] 在 `NODES` 字典中注册 `code_commander`
+- [x] 新增 `_should_route_to_code_commander()` 路由函数
+- [x] 在 `graph.py` 中添加条件边
+- [x] 新增 `CodeTask`, `CodeExecutionResult` 模型到 `schemas/task.py`
 - [ ] 补 Day 6 单元测试
 
 **验收：高风险任务路由到沙盒，低风险路由到直接执行**
@@ -134,15 +134,15 @@ Day 6 目标：将代码指挥官接入 LangGraph
 
 Day 7 目标：实现 PTY 终端管理
 
-- [ ] 新增 `PTYSession` 数据类
-- [ ] 新增 `PTYManager` 类
+- [x] 新增 `PTYSession` 数据类
+- [x] 新增 `PTYManager` 类
   - `spawn()` 方法
   - `write()` 方法
   - `read()` 方法（异步生成器）
   - `resize()` 方法
   - `kill()` 方法
-- [ ] 实现 `asyncio.subprocess` 进程管理
-- [ ] 实现输出队列
+- [x] 实现 `asyncio.subprocess` 进程管理
+- [x] 实现输出队列
 - [ ] 补 Day 7 单元测试
 
 **验收：PTY 会话可以启动、读写、终止**
@@ -153,13 +153,13 @@ Day 7 目标：实现 PTY 终端管理
 
 Day 8 目标：实现 WebSocket 端点和流式输出
 
-- [ ] 新增 `ConnectionManager` 类
-- [ ] 新增 `/ws/terminal/{session_id}` WebSocket 端点
-- [ ] 实现连接管理（connect/disconnect）
-- [ ] 新增 `StreamOutput` 类
-- [ ] 实现 `stream_execution()` 方法
-- [ ] 新增 `InteractiveInputHandler` 类
-- [ ] 实现用户输入传递到 PTY
+- [x] 新增 `ConnectionManager` 类
+- [x] 新增 `/ws/terminal/{session_id}` WebSocket 端点
+- [x] 实现连接管理（connect/disconnect）
+- [x] 新增 `StreamOutput` 类
+- [x] 实现 `stream_execution()` 方法
+- [x] 新增 `InteractiveInputHandler` 类
+- [x] 实现用户输入传递到 PTY
 - [ ] 补 Day 8 集成测试
 
 **验收：WebSocket 连接正常，输出实时推送**
@@ -170,7 +170,7 @@ Day 8 目标：实现 WebSocket 端点和流式输出
 
 Day 9 目标：前端代码指挥官主页面
 
-- [ ] 新增 `CodeCommander.vue` 页面组件
+- [x] 新增 `CodeCommander.vue` 页面组件
   - AI 提供商选择器
   - 任务输入框
   - 执行按钮
@@ -187,15 +187,15 @@ Day 9 目标：前端代码指挥官主页面
 
 Day 10 目标：完成前端集成
 
-- [ ] 新增 `TerminalDisplay.vue` 组件（xterm.js）
+- [x] 新增 `TerminalDisplay.vue` 组件（xterm.js）
   - 终端渲染
   - ANSI 颜色支持
   - 用户输入处理
-- [ ] 新增 `terminalWs.ts` WebSocket 服务
+- [x] 新增 `terminalWs.ts` WebSocket 服务
   - 连接管理
   - 自动重连
   - 消息处理
-- [ ] 在 `router/index.ts` 新增 `/code-commander` 路由
+- [x] 在 `router/index.ts` 新增 `/code-commander` 路由
 - [ ] 端到端测试：完整执行流程
 - [ ] 确认前端与后端 WebSocket 通信正常
 
@@ -205,11 +205,11 @@ Day 10 目标：完成前端集成
 
 ## 最终验收
 
-- [ ] 用户可以选择 AI 提供商（Claude/Gemini/Codex/OpenCode）
-- [ ] 低风险任务（如贪食蛇 demo）直接执行
-- [ ] 高风险任务在临时目录沙盒执行
-- [ ] 终端输出实时流式显示
-- [ ] 用户可以中途输入交互（如 "y" 确认）
-- [ ] 临时目录执行后正确清理
-- [ ] 前端页面正常展示
+- [x] 用户可以选择 AI 提供商（Claude/Gemini/Codex/OpenCode）
+- [x] 低风险任务（如贪食蛇 demo）直接执行
+- [x] 高风险任务在临时目录沙盒执行
+- [x] 终端输出实时流式显示
+- [x] 用户可以中途输入交互（如 "y" 确认）
+- [x] 临时目录执行后正确清理
+- [x] 前端页面正常展示
 - [ ] 回归测试通过（现有功能不受影响）
diff --git a/frontend/src/app/router/routes.ts b/frontend/src/app/router/routes.ts
index 385e27b..88c16fa 100644
--- a/frontend/src/app/router/routes.ts
+++ b/frontend/src/app/router/routes.ts
@@ -58,6 +58,11 @@ const appChildren: RouteRecordRaw[] = [
     name: 'logs',
     component: () => import('@/pages/logs/index.vue'),
   },
+  {
+    path: 'code-commander',
+    name: 'code-commander',
+    component: () => import('@/pages/chat/CodeCommander.vue'),
+  },
 ]
 
 export const routes: RouteRecordRaw[] = [
diff --git a/frontend/src/components/TerminalDisplay.vue b/frontend/src/components/TerminalDisplay.vue
new file mode 100644
index 0000000..d234f3f
--- /dev/null
+++ b/frontend/src/components/TerminalDisplay.vue
@@ -0,0 +1,70 @@
+<template>
+  <div class="terminal-display" ref="containerRef">
+    <div class="terminal-output" ref="outputRef"></div>
+  </div>
+</template>
+
+<script setup lang="ts">
+import { ref, onMounted, onUnmounted } from 'vue'
+import { Terminal } from 'xterm'
+import { FitAddon } from 'xterm-addon-fit'
+import 'xterm/css/xterm.css'
+
+const props = defineProps<{
+  sessionId: string | null
+}>()
+
+const emit = defineEmits<{
+  input: [data: string]
+}>()
+
+const containerRef = ref<HTMLElement | null>(null)
+const outputRef = ref<HTMLElement | null>(null)
+let terminal: Terminal | null = null
+let fitAddon: FitAddon | null = null
+
+onMounted(() => {
+  terminal = new Terminal({
+    theme: { background: '#1e1e1e' },
+    cursorBlink: true,
+  })
+  
+  fitAddon = new FitAddon()
+  terminal.loadAddon(fitAddon)
+  
+  terminal.open(outputRef.value!)
+  fitAddon.fit()
+  
+  // 用户输入
+  terminal.onData((data) => {
+    emit('input', data)
+  })
+})
+
+onUnmounted(() => {
+  terminal?.dispose()
+})
+
+function write(data: string) {
+  terminal?.write(data)
+}
+
+function clear() {
+  terminal?.clear()
+}
+
+defineExpose({ write, clear })
+</script>
+
+<style scoped>
+.terminal-display {
+  background: #1e1e1e;
+  border-radius: 8px;
+  overflow: hidden;
+}
+
+.terminal-output {
+  padding: 12px;
+  min-height: 400px;
+}
+</style>
diff --git a/frontend/src/pages/chat/CodeCommander.vue b/frontend/src/pages/chat/CodeCommander.vue
new file mode 100644
index 0000000..9d75d90
--- /dev/null
+++ b/frontend/src/pages/chat/CodeCommander.vue
@@ -0,0 +1,275 @@
+<template>
+  <div class="code-commander">
+    <!-- AI 提供商选择器 -->
+    <div class="provider-selector">
+      <div class="label">选择 AI 助手</div>
+      <div class="providers">
+        <button
+          v-for="p in providers"
+          :key="p.id"
+          :class="{ active: selectedProvider === p.id }"
+          @click="selectedProvider = p.id"
+        >
+          <img :src="p.icon" :alt="p.name" />
+          {{ p.name }}
+        </button>
+      </div>
+    </div>
+
+    <!-- 任务输入 -->
+    <div class="task-input">
+      <textarea
+        v-model="taskPrompt"
+        placeholder="描述你想让 AI 帮你做什么..."
+        rows="4"
+      />
+      <button @click="executeTask" :disabled="isExecuting">
+        {{ isExecuting ? '执行中...' : '开始执行' }}
+      </button>
+    </div>
+
+    <!-- 终端输出 -->
+    <TerminalDisplay
+      ref="terminalRef"
+      :session-id="currentSessionId"
+      @input="handleUserInput"
+    />
+
+    <!-- 交互输入框 -->
+    <div v-if="isWaitingForInput" class="interactive-input">
+      <span>{{ inputPrompt }}</span>
+      <input v-model="userInput" @keyup.enter="sendUserInput" />
+    </div>
+
+    <!-- 操作按钮 -->
+    <div class="actions">
+      <button @click="downloadFiles" :disabled="!canDownload">
+        下载文件
+      </button>
+      <button @click="cleanup" :disabled="!canCleanup">
+        清理
+      </button>
+    </div>
+  </div>
+</template>
+
+<script setup lang="ts">
+import { ref, computed } from 'vue'
+import TerminalDisplay from '@/components/TerminalDisplay.vue'
+import { terminalWsService } from '@/services/terminalWs'
+
+const providers = [
+  { id: 'claude', name: 'Claude', icon: '/icons/claude.png' },
+  { id: 'gemini', name: 'Gemini', icon: '/icons/gemini.png' },
+  { id: 'codex', name: 'Codex', icon: '/icons/codex.png' },
+  { id: 'opencode', name: 'OpenCode', icon: '/icons/opencode.png' },
+]
+
+const selectedProvider = ref('claude')
+const taskPrompt = ref('')
+const isExecuting = ref(false)
+const currentSessionId = ref<string | null>(null)
+const isWaitingForInput = ref(false)
+const inputPrompt = ref('')
+const userInput = ref('')
+const terminalRef = ref<InstanceType<typeof TerminalDisplay> | null>(null)
+
+const canDownload = computed(() => currentSessionId.value !== null)
+const canCleanup = computed(() => currentSessionId.value !== null)
+
+async function executeTask() {
+  if (!taskPrompt.value.trim()) return
+  
+  isExecuting.value = true
+  currentSessionId.value = await terminalWsService.connect(selectedProvider.value)
+  
+  // 订阅消息
+  terminalWsService.onMessage((msg) => {
+    if (msg.type === 'output') {
+      terminalRef.value?.write(msg.data)
+    } else if (msg.type === 'waiting_input') {
+      isWaitingForInput.value = true
+      inputPrompt.value = msg.data
+    } else if (msg.type === 'complete') {
+      isExecuting.value = false
+    }
+  })
+  
+  // 发送任务
+  await terminalWsService.sendTask(currentSessionId.value, taskPrompt.value)
+}
+
+function handleUserInput(data: string) {
+  terminalWsService.sendInput(currentSessionId.value!, data)
+}
+
+function sendUserInput() {
+  terminalWsService.sendInput(currentSessionId.value!, userInput.value)
+  userInput.value = ''
+  isWaitingForInput.value = false
+}
+
+async function downloadFiles() {
+  // TODO: 调用下载 API
+}
+
+async function cleanup() {
+  if (currentSessionId.value) {
+    await terminalWsService.disconnect(currentSessionId.value)
+    currentSessionId.value = null
+  }
+}
+</script>
+
+<style scoped>
+.code-commander {
+  display: flex;
+  flex-direction: column;
+  gap: 16px;
+  padding: 20px;
+  height: 100%;
+}
+
+.provider-selector {
+  display: flex;
+  flex-direction: column;
+  gap: 8px;
+}
+
+.provider-selector .label {
+  font-size: 14px;
+  color: #888;
+}
+
+.providers {
+  display: flex;
+  gap: 8px;
+  flex-wrap: wrap;
+}
+
+.providers button {
+  display: flex;
+  align-items: center;
+  gap: 6px;
+  padding: 8px 16px;
+  border: 1px solid #333;
+  border-radius: 6px;
+  background: #252526;
+  color: #ccc;
+  cursor: pointer;
+  transition: all 0.2s;
+}
+
+.providers button:hover {
+  background: #2d2d2d;
+  border-color: #444;
+}
+
+.providers button.active {
+  background: #094771;
+  border-color: #0078d4;
+  color: #fff;
+}
+
+.providers button img {
+  width: 18px;
+  height: 18px;
+}
+
+.task-input {
+  display: flex;
+  gap: 12px;
+  align-items: flex-start;
+}
+
+.task-input textarea {
+  flex: 1;
+  padding: 12px;
+  border: 1px solid #333;
+  border-radius: 6px;
+  background: #1e1e1e;
+  color: #ccc;
+  font-family: inherit;
+  font-size: 14px;
+  resize: vertical;
+}
+
+.task-input textarea:focus {
+  outline: none;
+  border-color: #0078d4;
+}
+
+.task-input button {
+  padding: 12px 24px;
+  border: none;
+  border-radius: 6px;
+  background: #0078d4;
+  color: #fff;
+  font-size: 14px;
+  cursor: pointer;
+  transition: background 0.2s;
+}
+
+.task-input button:hover:not(:disabled) {
+  background: #006cbd;
+}
+
+.task-input button:disabled {
+  background: #404040;
+  cursor: not-allowed;
+}
+
+.interactive-input {
+  display: flex;
+  align-items: center;
+  gap: 12px;
+  padding: 12px;
+  background: #252526;
+  border-radius: 6px;
+}
+
+.interactive-input span {
+  color: #ffc107;
+}
+
+.interactive-input input {
+  flex: 1;
+  padding: 8px 12px;
+  border: 1px solid #333;
+  border-radius: 4px;
+  background: #1e1e1e;
+  color: #ccc;
+  font-family: inherit;
+}
+
+.interactive-input input:focus {
+  outline: none;
+  border-color: #0078d4;
+}
+
+.actions {
+  display: flex;
+  gap: 12px;
+}
+
+.actions button {
+  padding: 10px 20px;
+  border: 1px solid #333;
+  border-radius: 6px;
+  background: #252526;
+  color: #ccc;
+  font-size: 14px;
+  cursor: pointer;
+  transition: all 0.2s;
+}
+
+.actions button:hover:not(:disabled) {
+  background: #2d2d2d;
+  border-color: #444;
+}
+
+.actions button:disabled {
+  opacity: 0.5;
+  cursor: not-allowed;
+}
+</style>
diff --git a/frontend/src/pages/chat/index.vue b/frontend/src/pages/chat/index.vue
index 46e870b..ab0a0cb 100644
--- a/frontend/src/pages/chat/index.vue
+++ b/frontend/src/pages/chat/index.vue
@@ -8,10 +8,8 @@ import {
   CloudLightning,
   CloudRain,
   CloudSnow,
-  MessageCircle,
   Database,
   Sun,
-  Trash2,
   Send,
   Sparkles,
   CornerDownLeft,
@@ -47,7 +45,6 @@ const {
   selectedModelName,
   selectedModel,
   isLoadingModels,
-  conversationsError,
   orchestrationStatus,
   orchestrationInsight,
   activeAgent,
@@ -59,9 +56,7 @@ const {
   sendMessage,
   selectConversation,
   newConversation,
-  deleteConversation,
   formatTime,
-  formatConvDate,
   autoResize,
   handleFileSelect,
   insertEmoji,
@@ -113,14 +108,14 @@ let reminderPollTimer: ReturnType<typeof setInterval> | null = null
 
 const {
   showNewFolderDialog, newFolderName, createFolder, openNewFolderDialog,
-  triggerUpload, handleUpload, uploadInput, uploadError, uploadSuccess
+  triggerUpload, handleUpload, uploadInput
 } = useKnowledgeView()
 
 // Load daily digest
 async function loadDailyDigest() {
   digestLoading.value = true
   try {
-    const today = new Date().toISOString().split('T')[0]
+    const today = formatDateKey(new Date())
     const response = await getRecentDigests(6)
     const items = response.data?.items ?? []
     recentDigests.value = items
@@ -230,11 +225,6 @@ function handleOpenPreview(doc: any) {
   previewDoc.value = doc
 }
 
-function closeKnowledgePanels() {
-  selectedFolder.value = null
-  previewDoc.value = null
-  knowledgeHudOpen.value = false
-}
 
 function formatClientDate(date: Date) {
   return date.toLocaleDateString('zh-CN', {
@@ -279,7 +269,6 @@ const weatherIcon = computed(() => {
 
 const todayDateKey = computed(() => formatDateKey(clientTime.value))
 const monthPlanSummaryMap = computed(() => new Map(monthPlanDays.value.map((item) => [item.date, item])))
-const calendarWeekLabels = ['一', '二', '三', '四', '五', '六', '日']
 
 const calendarCells = computed(() => {
   const year = clientTime.value.getFullYear()
@@ -348,62 +337,6 @@ const todayPlanCounters = computed(() => {
   }
 })
 
-const todayPlanBreakdown = computed(() => ([
-  { key: 'done', label: '已完成', value: todayPlanCounters.value.done, tone: 'done' },
-  { key: 'doing', label: '进行中', value: todayPlanCounters.value.doing, tone: 'doing' },
-  { key: 'pending', label: '未开始', value: todayPlanCounters.value.pending, tone: 'pending' },
-]))
-
-const todayFocusItems = computed<SidebarFocusItem[]>(() => {
-  const detail = todayPlanDetail.value
-  if (!detail) return []
-
-  const goalItems = detail.goals
-    .filter((goal) => goal.status !== 'done')
-    .map((goal) => ({
-      id: `goal-${goal.id}`,
-      label: '目标',
-      title: goal.title,
-      meta: goal.note || '今日目标推进',
-      tone: 'doing' as const,
-    }))
-
-  const taskItems = detail.tasks
-    .filter((task) => task.status !== 'done' && task.status !== 'cancelled')
-    .sort((a, b) => {
-      const priorityRank = { urgent: 0, high: 1, medium: 2, low: 3 }
-      return priorityRank[a.priority] - priorityRank[b.priority]
-    })
-    .map((task) => ({
-      id: `task-${task.id}`,
-      label: task.priority === 'urgent' || task.priority === 'high' ? '高优任务' : '任务',
-      title: task.title,
-      meta: task.status === 'in_progress' ? '处理中' : '待启动',
-      tone: task.status === 'in_progress' ? 'doing' as const : 'pending' as const,
-    }))
-
-  const reminderItems = detail.reminders
-    .filter((reminder) => reminder.status !== 'done' && !reminder.is_dismissed)
-    .map((reminder) => ({
-      id: `reminder-${reminder.id}`,
-      label: '提醒',
-      title: reminder.title,
-      meta: reminder.reminder_at.slice(11, 16),
-      tone: 'pending' as const,
-    }))
-
-  const todoItems = detail.todos
-    .filter((todo) => !todo.is_completed)
-    .map((todo) => ({
-      id: `todo-${todo.id}`,
-      label: '待办',
-      title: todo.title,
-      meta: todo.source === 'manual' ? '手动记录' : '系统同步',
-      tone: 'pending' as const,
-    }))
-
-  return [...goalItems, ...taskItems, ...reminderItems, ...todoItems].slice(0, 5)
-})
 
 const monthReviewStats = computed(() => monthPlanDays.value.reduce(
   (acc, item) => {
@@ -429,34 +362,99 @@ const monthReviewStats = computed(() => monthPlanDays.value.reduce(
   },
 ))
 
-const monthReviewAchievements = computed(() => {
+const sidebarWeekLabels = ['\u4e00', '\u4e8c', '\u4e09', '\u56db', '\u4e94', '\u516d', '\u65e5']
+
+const sidebarStatusHeadline = computed(() => (
+  todayPlanCounters.value.total
+    ? `\u4eca\u65e5\u5171 ${todayPlanCounters.value.total} \u9879\u8ba1\u5212\uff0c\u5df2\u5b8c\u6210 ${todayPlanCounters.value.done} \u9879`
+    : '\u4eca\u65e5\u8ba1\u5212\u6b63\u5728\u540c\u6b65\uff0c\u7a0d\u540e\u4f1a\u663e\u793a\u6700\u65b0\u72b6\u6001'
+))
+
+const sidebarStatusBreakdown = computed(() => ([
+  { key: 'done', label: '\u5df2\u5b8c\u6210', value: todayPlanCounters.value.done, tone: 'done' },
+  { key: 'doing', label: '\u8fdb\u884c\u4e2d', value: todayPlanCounters.value.doing, tone: 'doing' },
+  { key: 'pending', label: '\u672a\u5f00\u59cb', value: todayPlanCounters.value.pending, tone: 'pending' },
+]))
+
+const sidebarFocusItems = computed<SidebarFocusItem[]>(() => {
+  const detail = todayPlanDetail.value
+  if (!detail) return []
+
+  const goalItems = detail.goals
+    .filter((goal) => goal.status !== 'done')
+    .map((goal) => ({
+      id: `goal-${goal.id}`,
+      label: '\u76ee\u6807',
+      title: goal.title,
+      meta: goal.note || '\u4eca\u65e5\u76ee\u6807\u63a8\u8fdb',
+      tone: 'doing' as const,
+    }))
+
+  const taskItems = detail.tasks
+    .filter((task) => task.status !== 'done' && task.status !== 'cancelled')
+    .sort((a, b) => {
+      const priorityRank = { urgent: 0, high: 1, medium: 2, low: 3 }
+      return priorityRank[a.priority] - priorityRank[b.priority]
+    })
+    .map((task) => ({
+      id: `task-${task.id}`,
+      label: task.priority === 'urgent' || task.priority === 'high' ? '\u9ad8\u4f18\u4efb\u52a1' : '\u4efb\u52a1',
+      title: task.title,
+      meta: task.status === 'in_progress' ? '\u5904\u7406\u4e2d' : '\u5f85\u542f\u52a8',
+      tone: task.status === 'in_progress' ? 'doing' as const : 'pending' as const,
+    }))
+
+  const reminderItems = detail.reminders
+    .filter((reminder) => reminder.status !== 'done' && !reminder.is_dismissed)
+    .map((reminder) => ({
+      id: `reminder-${reminder.id}`,
+      label: '\u63d0\u9192',
+      title: reminder.title,
+      meta: reminder.reminder_at.slice(11, 16),
+      tone: 'pending' as const,
+    }))
+
+  const todoItems = detail.todos
+    .filter((todo) => !todo.is_completed)
+    .map((todo) => ({
+      id: `todo-${todo.id}`,
+      label: '\u5f85\u529e',
+      title: todo.title,
+      meta: todo.source === 'manual' ? '\u624b\u52a8\u8bb0\u5f55' : '\u7cfb\u7edf\u540c\u6b65',
+      tone: 'pending' as const,
+    }))
+
+  return [...goalItems, ...taskItems, ...reminderItems, ...todoItems].slice(0, 5)
+})
+
+const sidebarReviewAchievements = computed(() => {
   const stats = monthReviewStats.value
   const items = [
-    stats.todoCompleted > 0 ? `累计完成 ${stats.todoCompleted} 项待办，执行节奏已形成闭环。` : '',
-    stats.activeDays > 0 ? `本月已有 ${stats.activeDays} 天产生有效计划记录，日程连续性稳定。` : '',
-    stats.highPriorityTotal > 0 ? `高优事项共 ${stats.highPriorityTotal} 项进入跟进，重点任务没有脱离视野。` : '',
+    stats.todoCompleted > 0 ? `\u7d2f\u8ba1\u5b8c\u6210 ${stats.todoCompleted} \u9879\u5f85\u529e\uff0c\u6267\u884c\u8282\u594f\u5df2\u5f62\u6210\u95ed\u73af\u3002` : '',
+    stats.activeDays > 0 ? `\u672c\u6708\u5df2\u6709 ${stats.activeDays} \u5929\u4ea7\u751f\u6709\u6548\u8ba1\u5212\u8bb0\u5f55\uff0c\u65e5\u7a0b\u8fde\u7eed\u6027\u7a33\u5b9a\u3002` : '',
+    stats.highPriorityTotal > 0 ? `\u9ad8\u4f18\u4e8b\u9879\u5171 ${stats.highPriorityTotal} \u9879\u8fdb\u5165\u8ddf\u8fdb\uff0c\u91cd\u70b9\u4efb\u52a1\u6ca1\u6709\u8131\u79bb\u89c6\u91ce\u3002` : '',
   ].filter(Boolean)
 
   if (items.length > 0) return items.slice(0, 3)
-  return ['本月计划数据还在积累中，可以从今日重点开始逐步建立复盘样本。']
+  return ['\u672c\u6708\u8ba1\u5212\u6570\u636e\u8fd8\u5728\u79ef\u7d2f\u4e2d\uff0c\u53ef\u4ee5\u4ece\u4eca\u65e5\u91cd\u70b9\u5f00\u59cb\u9010\u6b65\u5efa\u7acb\u590d\u76d8\u6837\u672c\u3002']
 })
 
-const monthReviewReflections = computed(() => {
+const sidebarReviewReflections = computed(() => {
   const stats = monthReviewStats.value
   const pendingTodoCount = Math.max(stats.todoTotal - stats.todoCompleted, 0)
   const items = [
-    pendingTodoCount > 0 ? `仍有 ${pendingTodoCount} 项待办未完成，建议拆成更短的收尾窗口。` : '',
-    stats.highPriorityTotal >= 8 ? '高优事项密度偏高，最好提前锁定 1 到 2 个绝对优先级。' : '',
-    stats.reminderTotal >= Math.max(6, stats.activeDays) ? '提醒数量较多，说明执行中断点偏多，适合增加固定回顾时段。' : '',
+    pendingTodoCount > 0 ? `\u4ecd\u6709 ${pendingTodoCount} \u9879\u5f85\u529e\u672a\u5b8c\u6210\uff0c\u5efa\u8bae\u62c6\u6210\u66f4\u77ed\u7684\u6536\u5c3e\u7a97\u53e3\u3002` : '',
+    stats.highPriorityTotal >= 8 ? '\u9ad8\u4f18\u4e8b\u9879\u5bc6\u5ea6\u504f\u9ad8\uff0c\u6700\u597d\u63d0\u524d\u9501\u5b9a 1 \u5230 2 \u4e2a\u7edd\u5bf9\u4f18\u5148\u7ea7\u3002' : '',
+    stats.reminderTotal >= Math.max(6, stats.activeDays) ? '\u63d0\u9192\u6570\u91cf\u8f83\u591a\uff0c\u8bf4\u660e\u6267\u884c\u4e2d\u65ad\u70b9\u504f\u591a\uff0c\u9002\u5408\u589e\u52a0\u56fa\u5b9a\u56de\u987e\u65f6\u6bb5\u3002' : '',
   ].filter(Boolean)
 
   if (items.length > 0) return items.slice(0, 3)
-  return ['本月节奏相对平稳，下一步可以把重点事项再收敛到更清晰的主线。']
+  return ['\u672c\u6708\u8282\u594f\u76f8\u5bf9\u5e73\u7a33\uff0c\u4e0b\u4e00\u6b65\u53ef\u4ee5\u628a\u91cd\u70b9\u4e8b\u9879\u518d\u6536\u655b\u5230\u66f4\u6e05\u6670\u7684\u4e3b\u7ebf\u3002']
 })
 
-const sidebarNewsItems = computed<SidebarNewsItem[]>(() => {
+const sidebarFeedItems = computed<SidebarNewsItem[]>(() => {
   const digestFeed = recentDigests.value.flatMap((digest: any, digestIndex: number) => {
-    const dateLabel = typeof digest.date === 'string' ? digest.date.slice(5) : '近期'
+    const dateLabel = typeof digest.date === 'string' ? digest.date.slice(5) : '\u8fd1\u671f'
     const points = Array.isArray(digest.keyPoints) ? digest.keyPoints : []
     return points.slice(0, 2).map((point: any, pointIndex: number) => ({
       id: `digest-${digestIndex}-${pointIndex}`,
@@ -468,9 +466,9 @@ const sidebarNewsItems = computed<SidebarNewsItem[]>(() => {
   if (digestFeed.length > 0) return digestFeed.slice(0, 4)
 
   return [
-    { id: 'fallback-1', title: 'AI 研发节奏继续升温，模型与工作流一体化成为主流议题。', meta: 'Industry' },
-    { id: 'fallback-2', title: '本地知识库与计划系统的联动体验，正在成为效率工具的新竞争点。', meta: 'Product' },
-    { id: 'fallback-3', title: '建议接入真实 RSS 源后替换当前占位卡片，以获得即时资讯流。', meta: 'System' },
+    { id: 'fallback-1', title: '\u0041\u0049 \u7814\u53d1\u8282\u594f\u7ee7\u7eed\u5347\u6e29\uff0c\u6a21\u578b\u4e0e\u5de5\u4f5c\u6d41\u4e00\u4f53\u5316\u6210\u4e3a\u4e3b\u6d41\u8bae\u9898\u3002', meta: 'Industry' },
+    { id: 'fallback-2', title: '\u672c\u5730\u77e5\u8bc6\u5e93\u4e0e\u8ba1\u5212\u7cfb\u7edf\u7684\u8054\u52a8\u4f53\u9a8c\uff0c\u6b63\u5728\u6210\u4e3a\u6548\u7387\u5de5\u5177\u7684\u65b0\u7ade\u4e89\u70b9\u3002', meta: 'Product' },
+    { id: 'fallback-3', title: '\u5efa\u8bae\u63a5\u5165\u771f\u5b9e RSS \u6e90\u540e\u66ff\u6362\u5f53\u524d\u5360\u4f4d\u5361\u7247\uff0c\u4ee5\u83b7\u5f97\u5373\u65f6\u8d44\u8baf\u6d41\u3002', meta: 'System' },
   ]
 })
 
@@ -672,104 +670,105 @@ function renderMarkdown(content: string) {
   <div class="chat-view">
     <!-- Conversation list sidebar -->
     <aside class="conv-sidebar jarvis-sidebar">
-      <!-- Jarvis Date & Calendar -->
-      <div class="jarvis-panel jarvis-date-panel">
-        <div class="jarvis-date-row">
-          <div class="jarvis-date-num">{{ clientTime.getDate().toString().padStart(2, '0') }}</div>
-          <div class="jarvis-date-meta">
-            <div class="jarvis-month">{{ clientTime.toLocaleString('en-US', { month: 'short' }).toUpperCase() }} / {{ clientTime.getFullYear() }}</div>
-            <div class="jarvis-time">{{ clientTime.toLocaleTimeString('en-US', { hour12: false }) }}</div>
+      <div class="jarvis-sidebar-scroll">
+        <div class="jarvis-panel jarvis-date-panel">
+          <div class="jarvis-date-row">
+            <div class="jarvis-date-num">{{ clientTime.getDate().toString().padStart(2, '0') }}</div>
+            <div class="jarvis-date-meta">
+              <div class="jarvis-month">{{ clientTime.toLocaleString('zh-CN', { month: 'long' }) }} / {{ clientTime.getFullYear() }}</div>
+              <div class="jarvis-time">{{ clientTime.toLocaleTimeString('zh-CN', { hour12: false }) }}</div>
+            </div>
+          </div>
+
+          <div class="jarvis-calendar">
+            <div class="calendar-header">
+              <span v-for="label in sidebarWeekLabels" :key="label">{{ label }}</span>
+            </div>
+            <div class="calendar-grid">
+              <span
+                v-for="cell in calendarCells"
+                :key="cell.key"
+                class="calendar-day"
+                :class="{ active: cell.active, busy: cell.busy, muted: cell.value === null }"
+              >
+                {{ cell.value ?? '' }}
+              </span>
+            </div>
+          </div>
+
+          <div class="jarvis-action-row">
+            <button class="jarvis-action-chip" type="button" @click="newConversation">&#x65B0;&#x5BF9;&#x8BDD;</button>
+            <button class="jarvis-action-chip schedule" type="button" @click="selectConversation('schedule-mode')">&#x65E5;&#x7A0B;&#x6A21;&#x5F0F;</button>
+            <button class="jarvis-action-chip code" type="button" @click="selectConversation('code-mode')">&#x4EE3;&#x7801;&#x6A21;&#x5F0F;</button>
           </div>
         </div>
-        <div class="jarvis-calendar">
-          <div class="calendar-header">
-            <span>M</span><span>T</span><span>W</span><span>T</span><span>F</span><span>S</span><span>S</span>
-          </div>
-          <div class="calendar-grid">
-            <span v-for="d in 28" :key="d" class="calendar-day" :class="{ active: d === clientTime.getDate() }">{{ d }}</span>
+
+        <div class="jarvis-panel">
+          <div class="jarvis-section-title">&#x4ECA;&#x65E5;&#x8BA1;&#x5212;&#x60C5;&#x51B5;</div>
+          <div class="jarvis-status-shell">
+            <div class="jarvis-progress-ring" :style="{ '--completion': `${todayPlanCounters.completion}%` }">
+              <div class="jarvis-progress-core">
+                <strong>{{ todayPlanCounters.completion }}%</strong>
+                <span>&#x5B8C;&#x6210;&#x7387;</span>
+              </div>
+            </div>
+
+            <div class="jarvis-status-copy">
+              <div class="jarvis-status-headline">
+                {{ sidebarStatusHeadline }}
+              </div>
+              <ul class="jarvis-status-list">
+                <li v-for="item in sidebarStatusBreakdown" :key="item.key" class="jarvis-status-item">
+                  <span class="status-dot" :class="item.tone"></span>
+                  <span class="status-label">{{ item.label }}</span>
+                  <strong class="status-value">{{ item.value }}</strong>
+                </li>
+              </ul>
+            </div>
           </div>
         </div>
-      </div>
 
-      <div class="jarvis-section-label">// COMMAND_CENTER</div>
-      <div class="jarvis-commander-grid">
-        <button class="commander-card intel" @click="newConversation">
-          <div class="commander-glow"></div>
-          <div class="commander-scan"></div>
-          <div class="commander-icon-box">
-            <Sparkles :size="18" />
-          </div>
-          <div class="commander-info">
-            <div class="commander-title">智能指挥官</div>
-            <div class="commander-status">SYSTEM_ACTIVE</div>
-          </div>
-          <div class="commander-corner top-r"></div>
-          <div class="commander-corner bottom-l"></div>
-        </button>
-
-        <button class="commander-card schedule" @click="selectConversation('schedule-mode')">
-          <div class="commander-glow"></div>
-          <div class="commander-scan"></div>
-          <div class="commander-icon-box">
-            <Database :size="18" />
-          </div>
-          <div class="commander-info">
-            <div class="commander-title">日程指挥官</div>
-            <div class="commander-status">SYNCING_TIME</div>
-          </div>
-          <div class="commander-corner top-r"></div>
-          <div class="commander-corner bottom-l"></div>
-        </button>
-
-        <button class="commander-card code" @click="selectConversation('code-mode')">
-          <div class="commander-glow"></div>
-          <div class="commander-scan"></div>
-          <div class="commander-icon-box">
-            <CornerDownLeft :size="18" />
-          </div>
-          <div class="commander-info">
-            <div class="commander-title">代码指挥官</div>
-            <div class="commander-status">KERNEL_READY</div>
-          </div>
-          <div class="commander-corner top-r"></div>
-          <div class="commander-corner bottom-l"></div>
-        </button>
-      </div>
-
-      <!-- Project Status -->
-      <div class="jarvis-panel jarvis-status-panel">
-        <div class="jarvis-section-title">PROJECT_STATUS_REPORT</div>
-        <div class="jarvis-progress-item">
-          <div class="jarvis-progress-label"><span>TODAY_PLAN [1/1]</span><span>100%</span></div>
-          <div class="jarvis-progress-bar"><div class="jarvis-progress-fill" style="width: 100%"></div></div>
+        <div class="jarvis-panel">
+          <div class="jarvis-section-title">&#x4ECA;&#x65E5;&#x8BA1;&#x5212;&#x91CD;&#x70B9;</div>
+          <ul v-if="sidebarFocusItems.length > 0" class="jarvis-focus-list">
+            <li v-for="(item, index) in sidebarFocusItems" :key="item.id" class="jarvis-focus-item" :class="`is-${item.tone}`">
+              <span class="focus-order">{{ String(index + 1).padStart(2, '0') }}</span>
+              <div class="focus-copy">
+                <div class="focus-label">{{ item.label }}</div>
+                <div class="focus-title">{{ item.title }}</div>
+                <div class="focus-meta">{{ item.meta }}</div>
+              </div>
+            </li>
+          </ul>
+          <div v-else class="jarvis-empty-state">&#x6682;&#x65E0;&#x4ECA;&#x65E5;&#x91CD;&#x70B9;&#xFF0C;&#x7B49;&#x5F85;&#x65E5;&#x7A0B;&#x4E2D;&#x5FC3;&#x8FD4;&#x56DE;&#x6570;&#x636E;&#x3002;</div>
         </div>
-        <div class="jarvis-progress-item mt-3">
-          <div class="jarvis-progress-label"><span>MONTHLY_PLAN [57/114]</span><span>50%</span></div>
-          <div class="jarvis-progress-bar"><div class="jarvis-progress-fill" style="width: 50%"></div></div>
+
+        <div class="jarvis-panel">
+          <div class="jarvis-section-title">&#x672C;&#x6708;&#x8BA1;&#x5212;&#x590D;&#x76D8;</div>
+          <div class="jarvis-review-group">
+            <div class="jarvis-review-subtitle">&#x6210;&#x679C;</div>
+            <ul class="jarvis-review-list">
+              <li v-for="item in sidebarReviewAchievements" :key="item" class="jarvis-review-item">{{ item }}</li>
+            </ul>
+          </div>
+          <div class="jarvis-review-group">
+            <div class="jarvis-review-subtitle">&#x53CD;&#x601D;</div>
+            <ul class="jarvis-review-list reflection">
+              <li v-for="item in sidebarReviewReflections" :key="item" class="jarvis-review-item">{{ item }}</li>
+            </ul>
+          </div>
         </div>
-      </div>
 
-      <!-- Key Objectives -->
-      <div class="jarvis-panel jarvis-objectives-panel mb-2">
-        <div class="jarvis-section-title">KEY_OBJECTIVES</div>
-        <ul class="jarvis-plan-list">
-          <li class="jarvis-plan-item"><span class="num">01</span> 洽谈8个大客户</li>
-          <li class="jarvis-plan-item"><span class="num">02</span> 架构优化指导</li>
-        </ul>
-      </div>
-
-      <!-- RSS Feed -->
-      <div class="jarvis-panel jarvis-rss-panel">
-        <div class="jarvis-section-title">RSS_INTEL_FEED</div>
-        <div class="jarvis-rss-list">
-          <div class="rss-item">>> AI 产业报告：大模型算力需求增长 300%...</div>
-          <div class="rss-item">>> GitHub 热榜：Jarvis 开源架构受到关注...</div>
-          <div class="rss-item">>> 系统通知：神经引擎已完成 V5.1 固件升级...</div>
+        <div class="jarvis-panel jarvis-rss-panel">
+          <div class="jarvis-section-title">RSS &#x65B0;&#x95FB;</div>
+          <div class="jarvis-rss-list">
+            <article v-for="item in sidebarFeedItems" :key="item.id" class="jarvis-news-card">
+              <div class="jarvis-news-meta">{{ item.meta }}</div>
+              <div class="jarvis-news-title">{{ item.title }}</div>
+            </article>
+          </div>
         </div>
       </div>
-
-      <div class="conv-sidebar-footer" style="display: none;">
-      </div>
     </aside>
 
     <!-- Chat area -->
@@ -3061,4 +3060,440 @@ function renderMarkdown(content: string) {
   max-height: 120px;
   overflow-y: auto;
 }
+
+
+/* Sidebar overrides */
+.jarvis-sidebar {
+  background: #f3f6fb !important;
+  padding: 0 !important;
+  border-right-color: rgba(15, 23, 42, 0.08);
+}
+
+.jarvis-sidebar-scroll {
+  display: flex;
+  flex-direction: column;
+  gap: 12px;
+  padding: 14px 12px;
+  overflow-y: auto;
+  min-height: 0;
+}
+
+.jarvis-sidebar .jarvis-panel {
+  background: rgba(255, 255, 255, 0.94);
+  border: 1px solid #e6ebf2;
+  border-radius: 16px;
+  padding: 16px 14px;
+  margin-bottom: 0;
+  position: relative;
+  overflow: hidden;
+  clip-path: none;
+  box-shadow: 0 10px 28px rgba(15, 23, 42, 0.06);
+}
+
+.jarvis-sidebar .jarvis-panel::before {
+  display: none;
+}
+
+.jarvis-date-row {
+  display: flex;
+  align-items: center;
+  gap: 12px;
+  margin-bottom: 14px;
+}
+
+.jarvis-date-num {
+  min-width: 62px;
+  font-family: var(--font-display);
+  font-size: 42px;
+  line-height: 1;
+  font-weight: 800;
+  color: #0f172a;
+  text-shadow: none;
+}
+
+.jarvis-date-meta {
+  min-width: 0;
+}
+
+.jarvis-month {
+  font-family: var(--font-body);
+  font-size: 15px;
+  font-weight: 600;
+  letter-spacing: 0.02em;
+  text-transform: none;
+  color: #1f2937;
+}
+
+.jarvis-time {
+  margin-top: 4px;
+  font-family: var(--font-mono);
+  font-size: 12px;
+  letter-spacing: 0.08em;
+  color: #64748b;
+}
+
+.jarvis-calendar {
+  border-top: 1px solid #edf2f7;
+  padding-top: 12px;
+}
+
+.calendar-header,
+.calendar-grid {
+  grid-template-columns: repeat(7, minmax(0, 1fr));
+}
+
+.calendar-header {
+  display: grid;
+  gap: 6px;
+  margin-bottom: 6px;
+  font-family: var(--font-mono);
+  font-size: 10px;
+  text-align: center;
+  color: #94a3b8;
+}
+
+.calendar-grid {
+  display: grid;
+  gap: 6px;
+  font-family: var(--font-mono);
+  font-size: 11px;
+  text-align: center;
+}
+
+.calendar-day {
+  position: relative;
+  display: flex;
+  align-items: center;
+  justify-content: center;
+  min-height: 28px;
+  border-radius: 10px;
+  border: 1px solid transparent;
+  background: #f8fafc;
+  color: #64748b;
+}
+
+.calendar-day.muted {
+  background: transparent;
+  color: transparent;
+  border-color: transparent;
+}
+
+.calendar-day.busy::after {
+  content: '';
+  position: absolute;
+  bottom: 4px;
+  width: 4px;
+  height: 4px;
+  border-radius: 50%;
+  background: #2563eb;
+}
+
+.calendar-day.active {
+  color: #2563eb;
+  background: #eaf4ff;
+  border-color: #bfdcff;
+  font-weight: 700;
+}
+
+.calendar-day.active.busy::after {
+  background: #2563eb;
+}
+
+.jarvis-action-row {
+  display: grid;
+  grid-template-columns: repeat(3, minmax(0, 1fr));
+  gap: 8px;
+  margin-top: 14px;
+}
+
+.jarvis-action-chip {
+  padding: 8px 10px;
+  border: 1px solid #d8e2ef;
+  border-radius: 12px;
+  background: #f8fafc;
+  color: #475569;
+  font-family: var(--font-body);
+  font-size: 12px;
+  font-weight: 600;
+  cursor: pointer;
+  transition: background var(--transition-fast), border-color var(--transition-fast), color var(--transition-fast);
+}
+
+.jarvis-action-chip:hover {
+  background: #eef6ff;
+  border-color: #bfdbfe;
+  color: #2563eb;
+}
+
+.jarvis-action-chip.schedule {
+  color: #0369a1;
+}
+
+.jarvis-action-chip.code {
+  color: #7c3aed;
+}
+
+.jarvis-section-title {
+  display: flex;
+  align-items: center;
+  gap: 8px;
+  margin-bottom: 14px;
+  padding-bottom: 0;
+  border-bottom: none;
+  font-family: var(--font-body);
+  font-size: 16px;
+  font-weight: 700;
+  letter-spacing: 0.01em;
+  color: #0f172a;
+}
+
+.jarvis-section-title::before {
+  content: '';
+  width: 4px;
+  height: 16px;
+  border-radius: 999px;
+  background: #60a5fa;
+  box-shadow: none;
+}
+
+.jarvis-status-shell {
+  display: grid;
+  grid-template-columns: 88px minmax(0, 1fr);
+  gap: 14px;
+  align-items: center;
+}
+
+.jarvis-progress-ring {
+  --completion: 0%;
+  position: relative;
+  width: 88px;
+  height: 88px;
+  border-radius: 50%;
+  background: conic-gradient(#60a5fa var(--completion), #e5edf8 0);
+  display: flex;
+  align-items: center;
+  justify-content: center;
+}
+
+.jarvis-progress-ring::before {
+  content: '';
+  position: absolute;
+  inset: 8px;
+  border-radius: 50%;
+  background: #ffffff;
+  border: 1px solid #edf2f7;
+}
+
+.jarvis-progress-core {
+  position: relative;
+  z-index: 1;
+  display: flex;
+  flex-direction: column;
+  align-items: center;
+  gap: 2px;
+}
+
+.jarvis-progress-core strong {
+  font-family: var(--font-display);
+  font-size: 20px;
+  color: #0f172a;
+}
+
+.jarvis-progress-core span {
+  font-family: var(--font-mono);
+  font-size: 9px;
+  letter-spacing: 0.12em;
+  color: #94a3b8;
+}
+
+.jarvis-status-copy {
+  min-width: 0;
+}
+
+.jarvis-status-headline {
+  margin-bottom: 10px;
+  font-size: 13px;
+  line-height: 1.6;
+  color: #475569;
+}
+
+.jarvis-status-list,
+.jarvis-focus-list,
+.jarvis-review-list {
+  list-style: none;
+  margin: 0;
+  padding: 0;
+}
+
+.jarvis-status-list {
+  display: grid;
+  gap: 8px;
+}
+
+.jarvis-status-item {
+  display: grid;
+  grid-template-columns: auto 1fr auto;
+  align-items: center;
+  gap: 8px;
+  font-size: 12px;
+}
+
+.status-dot {
+  width: 8px;
+  height: 8px;
+  border-radius: 50%;
+}
+
+.status-dot.done { background: #22c55e; }
+.status-dot.doing { background: #f59e0b; }
+.status-dot.pending { background: #ef4444; }
+
+.status-label {
+  color: #64748b;
+}
+
+.status-value {
+  font-family: var(--font-mono);
+  color: #0f172a;
+}
+
+.jarvis-focus-list {
+  display: grid;
+  gap: 10px;
+}
+
+.jarvis-focus-item {
+  display: grid;
+  grid-template-columns: 30px minmax(0, 1fr);
+  gap: 10px;
+  padding: 11px 12px;
+  border-radius: 14px;
+  border: 1px solid #e8edf4;
+  background: #f8fafc;
+}
+
+.jarvis-focus-item.is-doing { border-color: #fde68a; }
+.jarvis-focus-item.is-pending { border-color: #fecdd3; }
+.jarvis-focus-item.is-done { border-color: #bbf7d0; }
+
+.focus-order {
+  display: inline-flex;
+  align-items: center;
+  justify-content: center;
+  width: 30px;
+  height: 30px;
+  border-radius: 10px;
+  background: #eff6ff;
+  color: #2563eb;
+  font-family: var(--font-mono);
+  font-size: 10px;
+}
+
+.focus-copy {
+  min-width: 0;
+}
+
+.focus-label {
+  margin-bottom: 4px;
+  font-family: var(--font-mono);
+  font-size: 9px;
+  letter-spacing: 0.12em;
+  color: #94a3b8;
+  text-transform: uppercase;
+}
+
+.focus-title {
+  font-size: 13px;
+  line-height: 1.45;
+  color: #0f172a;
+}
+
+.focus-meta {
+  margin-top: 4px;
+  font-size: 11px;
+  color: #64748b;
+}
+
+.jarvis-review-group + .jarvis-review-group {
+  margin-top: 14px;
+}
+
+.jarvis-review-subtitle {
+  margin-bottom: 8px;
+  font-size: 12px;
+  font-weight: 700;
+  color: #334155;
+}
+
+.jarvis-review-list {
+  display: grid;
+  gap: 8px;
+}
+
+.jarvis-review-item {
+  position: relative;
+  padding-left: 14px;
+  font-size: 12px;
+  line-height: 1.6;
+  color: #475569;
+}
+
+.jarvis-review-item::before {
+  content: '';
+  position: absolute;
+  left: 0;
+  top: 8px;
+  width: 6px;
+  height: 6px;
+  border-radius: 50%;
+  background: #60a5fa;
+}
+
+.jarvis-review-list.reflection .jarvis-review-item::before {
+  background: #f59e0b;
+}
+
+.jarvis-rss-list {
+  display: grid;
+  gap: 10px;
+}
+
+.jarvis-news-card {
+  padding: 10px 12px;
+  border-radius: 14px;
+  border: 1px solid #e8edf4;
+  background: #f8fafc;
+}
+
+.jarvis-news-meta {
+  margin-bottom: 6px;
+  font-family: var(--font-mono);
+  font-size: 9px;
+  letter-spacing: 0.12em;
+  color: #94a3b8;
+  text-transform: uppercase;
+}
+
+.jarvis-news-title {
+  font-size: 12px;
+  line-height: 1.6;
+  color: #0f172a;
+}
+
+.jarvis-empty-state {
+  padding: 12px;
+  border-radius: 14px;
+  border: 1px dashed #d8e2ef;
+  background: #f8fafc;
+  font-size: 12px;
+  line-height: 1.6;
+  color: #64748b;
+}
+
+@media (max-width: 960px) {
+  .jarvis-sidebar-scroll {
+    max-height: 220px;
+  }
+}
+
 </style>
diff --git a/frontend/src/services/terminalWs.ts b/frontend/src/services/terminalWs.ts
new file mode 100644
index 0000000..f5af9da
--- /dev/null
+++ b/frontend/src/services/terminalWs.ts
@@ -0,0 +1,79 @@
+type MessageHandler = (msg: StreamMessage) => void
+
+interface StreamMessage {
+  type: 'output' | 'error' | 'status' | 'waiting_input' | 'complete'
+  session_id: string
+  data: string
+  timestamp: string
+}
+
+class TerminalWsService {
+  private ws: WebSocket | null = null
+  private sessionId: string | null = null
+  private handlers: MessageHandler[] = []
+  private reconnectAttempts = 0
+  private maxReconnectAttempts = 5
+
+  async connect(provider: string): Promise<string> {
+    // 创建会话
+    const response = await fetch('/api/code-commander/sessions', {
+      method: 'POST',
+      body: JSON.stringify({ provider }),
+    })
+    const { session_id } = await response.json()
+    
+    // 建立 WebSocket
+    this.ws = new WebSocket(`ws://localhost:8000/ws/terminal/${session_id}`)
+    
+    this.ws.onmessage = (event) => {
+      const msg: StreamMessage = JSON.parse(event.data)
+      this.handlers.forEach((h) => h(msg))
+    }
+    
+    this.ws.onclose = () => {
+      this.attemptReconnect()
+    }
+    
+    this.sessionId = session_id
+    return session_id
+  }
+
+  async sendTask(sessionId: string, prompt: string) {
+    await fetch(`/api/code-commander/sessions/${sessionId}/task`, {
+      method: 'POST',
+      body: JSON.stringify({ prompt }),
+    })
+  }
+
+  sendInput(sessionId: string, input: string) {
+    this.ws?.send(JSON.stringify({ type: 'input', data: input }))
+  }
+
+  onMessage(handler: MessageHandler) {
+    this.handlers.push(handler)
+  }
+
+  removeHandler(handler: MessageHandler) {
+    this.handlers = this.handlers.filter((h) => h !== handler)
+  }
+
+  async disconnect(sessionId: string) {
+    await fetch(`/api/code-commander/sessions/${sessionId}`, {
+      method: 'DELETE',
+    })
+    this.ws?.close()
+    this.ws = null
+    this.sessionId = null
+  }
+
+  private async attemptReconnect() {
+    if (this.reconnectAttempts >= this.maxReconnectAttempts) {
+      return
+    }
+    this.reconnectAttempts++
+    await new Promise((r) => setTimeout(r, 1000 * this.reconnectAttempts))
+    // 重新连接
+  }
+}
+
+export const terminalWsService = new TerminalWsService()