feat(outlook): 添加Outlook集成功能支持

添加完整的Outlook MCP集成，包括邮件和日历功能，通过AuthZ模式进行认证和权限管理，
支持邮箱连接、断开、状态检查和数据同步等功能。

fix(config): 统一配置文件路径从.nanobot到.beaver

将配置文件路径从/root/.nanobot统一更改为/root/.beaver，更新Dockerfile中的环境变量定义，
确保所有组件使用一致的配置目录结构。

feat(agent): 添加代理删除功能和助手身份提示

为代理注册表添加delete_agent方法，实现代理的动态删除功能；同时添加海狸助手身份提示，
确保AI助手在交互中保持一致的身份认知。

feat(engine): 增强引擎循环并添加意图决策快照

扩展AgentLoop类，添加intent_agent_decision参数用于意图驱动的代理决策，并在会话中记录
决策快照，便于后续分析和调试。

feat(authz): 扩展认证客户端功能

为AuthzClient添加设置权限、用户注册、后端注册和Outlook设置管理等新方法，增强系统
的认证和授权能力。

app-instance/backend/beaver/skills/builtin/intent-agent-router/SKILL.md

@@ -0,0 +1,59 @@
+---
+name: intent-agent-router
+description: Internal routing guidance for the first-layer Intent Agent.
+internal: true
+---
+# Intent Agent Router
+
+## Role
+
+You are the first-layer Intent Agent. You are not the main assistant and you do not execute the user request.
+
+Your only job is to classify the current user message into one routing decision:
+
+- `simple_chat`
+- `continue_task`
+- `new_task`
+- `close_task`
+- `abandon_task`
+
+Return compact JSON only. Never answer the user directly.
+
+## Core Boundary
+
+Choose `simple_chat` only when the message can be answered safely from ordinary language understanding without tools, external data, local files, user-private data, execution, validation, or multi-step work.
+
+Choose `new_task` when the user asks for anything that needs the main Task agent's capabilities, including tools, skills, files, web/search, execution, iteration, planning, validation, or multi-agent work.
+
+The Intent Agent has no tools. If a request needs a tool, do not apologize and do not say you cannot access it. Route it to Task mode so the main agent can use tools.
+
+## Must Create Task
+
+Choose `new_task` when there is no active task and the request asks to:
+
+- look up, search, browse, fetch, verify, check, monitor, compare, or summarize current/external information
+- answer about today's weather, live conditions, latest news, prices, schedules, exchange rates, regulations, releases, or other changing facts
+- inspect, read, write, patch, run, test, build, deploy, debug, or modify local files or systems
+- use email, calendar, messages, databases, MCP tools, shell commands, web APIs, or other integrations
+- produce a deliverable that needs multiple steps, validation, or follow-up execution
+
+Examples that must be `new_task`:
+
+- "帮我查一下今天珠海天气"
+- "查一下最新的 OpenAI API 价格"
+- "看看这个项目测试为什么失败"
+- "帮我改一下登录页面"
+- "给我查一下明天的航班"
+
+## Simple Chat
+
+Choose `simple_chat` only for lightweight conversation or stable knowledge that does not need tools.
+
+Examples:
+
+- "你好"
+- "解释一下什么是递归"
+- "把这句话润色一下"
+- "给我一个学习 Python 的大纲"
+
+If uncertain whether tools may be needed, prefer `new_task`.

app-instance/backend/beaver/skills/catalog/loader.py

@@ -67,7 +67,12 @@ class SkillsLoader:
         self.extra_dirs = [Path(item) for item in (extra_dirs or [])]
         self.skill_store = skill_store or SkillSpecStore(self.workspace)
 
-    def list_skills(self, *, filter_unavailable: bool = True) -> list[SkillRecord]:
+    def list_skills(
+        self,
+        *,
+        filter_unavailable: bool = True,
+        include_internal: bool = False,
+    ) -> list[SkillRecord]:
         """列出当前可见的 skills。
 
         优先级：
@@ -80,9 +85,11 @@ class SkillsLoader:
 
         found: dict[str, SkillRecord] = {}
 
-        for record in self.list_published_skills():
+        for record in self.list_published_skills(filter_unavailable=filter_unavailable):
             if record.name in found:
                 continue
+            if not include_internal and self._record_internal(record):
+                continue
             if filter_unavailable and not self._record_available(record):
                 continue
             found[record.name] = record
@@ -101,6 +108,8 @@ class SkillsLoader:
                 if name in found:
                     continue
                 frontmatter, body = parse_frontmatter(skill_file.read_text(encoding="utf-8"))
+                if not include_internal and _truthy(frontmatter.get("internal")):
+                    continue
                 normalized_frontmatter = dict(frontmatter)
                 record = SkillRecord(
                     name=name,
@@ -375,11 +384,15 @@ class SkillsLoader:
         return []
 
     def _find_record(self, name: str) -> SkillRecord | None:
-        for record in self.list_skills(filter_unavailable=False):
+        for record in self.list_skills(filter_unavailable=False, include_internal=True):
             if record.name == name:
                 return record
         return None
 
+    @staticmethod
+    def _record_internal(record: SkillRecord) -> bool:
+        return _truthy((record.frontmatter or {}).get("internal"))
+
     def _record_available(self, record: SkillRecord) -> bool:
         content = record.path.read_text(encoding="utf-8")
         frontmatter, _ = parse_frontmatter(content)
@@ -405,3 +418,9 @@ class SkillsLoader:
 def summarize_body(body: str) -> str:
     cleaned = " ".join(line.strip() for line in body.splitlines()[:3] if line.strip()).strip()
     return cleaned[:240]
+
+
+def _truthy(value: Any) -> bool:
+    if isinstance(value, bool):
+        return value
+    return str(value or "").strip().lower() in {"1", "true", "yes", "y", "on"}

app-instance/backend/beaver/skills/learning/service.py

@@ -181,7 +181,7 @@ class SkillLearningService:
         if candidate.kind == "new_skill":
             payload = await self.synthesizer.synthesize_new_skill(candidate, packet, provider, model)
             return self.draft_service.create_new_skill_draft(
-                skill_name=self._suggest_skill_name(candidate, packet),
+                skill_name=self._suggest_skill_name(candidate, packet, payload.get("frontmatter")),
                 proposed_content=payload["content"],
                 proposed_frontmatter=payload["frontmatter"],
                 created_by="learning-loop",
@@ -382,15 +382,34 @@ class SkillLearningService:
         return " ".join(words[:8]).strip()
 
     @staticmethod
-    def _suggest_skill_name(candidate: SkillLearningCandidate, packet: EvidencePacket) -> str:
+    def _suggest_skill_name(
+        candidate: SkillLearningCandidate,
+        packet: EvidencePacket,
+        frontmatter: dict[str, Any] | None = None,
+    ) -> str:
         if candidate.related_skill_names:
             return candidate.related_skill_names[0]
-        if packet.task_summaries:
-            seed = re.sub(r"[^a-z0-9]+", "-", packet.task_summaries[0].lower()).strip("-")
+        if isinstance(frontmatter, dict):
+            description = str(frontmatter.get("description") or "")
+            seed = SkillLearningService._slugify_skill_name(description)
             if seed:
-                return seed[:48]
+                return seed
+        if packet.task_summaries:
+            seed = SkillLearningService._slugify_skill_name(packet.task_summaries[0])
+            if seed:
+                return seed
         return f"generated-skill-{uuid4().hex[:8]}"
 
+    @staticmethod
+    def _slugify_skill_name(value: str) -> str:
+        seed = re.sub(r"[^a-z0-9]+", "-", value.lower()).strip("-")
+        seed = re.sub(r"-+", "-", seed)
+        if not seed or seed.isdigit() or len(seed) < 3:
+            return ""
+        words = [part for part in seed.split("-") if part and not part.isdigit()]
+        seed = "-".join(words) or seed
+        return seed[:48].strip("-")
+
     @staticmethod
     def _parse_timestamp(value: str) -> datetime | None:
         try: