diff --git a/PLUGINS/ClaudeCode/skills/asp-module-creator-zh/DESC.md b/PLUGINS/ClaudeCode/skills/asp-module-creator-zh/DESC.md
new file mode 100644
index 0000000..e69de29
diff --git a/PLUGINS/ClaudeCode/skills/asp-module-creator-zh/SKILL.md b/PLUGINS/ClaudeCode/skills/asp-module-creator-zh/SKILL.md
new file mode 100644
index 0000000..3f08f3d
--- /dev/null
+++ b/PLUGINS/ClaudeCode/skills/asp-module-creator-zh/SKILL.md
@@ -0,0 +1,210 @@
+---
+name: asp-siem-zh
+description: '用于在 ASP SIEM 中进行日志调查、事件检索、字段探索和结构化分析，适合从模糊线索到精确验证的调查任务。'
+argument-hint: 'explore schema [index] | search <keyword> from <UTC start> to <UTC end> | adaptive query <index_name> <time range> [filters] [aggregations]'
+compatibility: connect to asp mcp server
+metadata:
+  author: Funnywolf
+  version: 0.3.0
+  mcp-server: asp
+  category: cyber security
+  tags: [ SIEM, search, SOC, hunting, investigation ]
+  documentation: https://asp.viperrtp.com/
+---
+
+# ASP SIEM
+
+当用户要在 ASP 中进行 SIEM 调查时，使用这个 skill。重点是帮助用户根据调查目标选择合适的查询方式，并输出对分析有价值的证据。
+
+## 适用场景
+
+- 用户要在 ASP SIEM 中查找、确认或补充安全事件证据。
+- 用户想先了解有哪些索引、字段或可用数据源。
+- 用户想根据 IOC、告警上下文或关键词搜索相关日志。
+- 用户想在已知范围内做进一步过滤、统计或结构化分析。
+- 用户想从 alert、artifact 或 case 继续向 SIEM 侧取证。
+
+## 运行规则
+
+- 如果用户请求已经隐含 SIEM 搜索，不要反问用户要选哪种操作。
+- 只收集所选路径缺失的必要输入。
+- 当用户不知道正确索引或字段时，使用 `siem_explore_schema`。
+- `siem_keyword_search` 适合关键词驱动的搜索，既可用于广范围探索，也可用于已知 index 下的快速检索。
+- `siem_adaptive_query` 适合精确字段过滤、受控聚合和稳定复现，能力上更接近一个结构化 SIEM API。
+- 如果用户给出相对时间窗口，先调用 `get_current_time`，从返回的本地时间加时区推导出可用的 UTC 范围。
+- 优化目标是有用证据，而不是最大原始输出。
+
+
+## 函数说明
+
+### `siem_keyword_search`
+
+适用场景：
+
+- 用户手里有 IP、域名、用户名、主机名、hash、进程名、邮件地址、报错文本或其他关键词，想直接查有没有相关日志。
+- 用户还不确定日志在哪个 index，想先在较大范围或全局范围搜索。
+- 用户想像 SIEM 检索界面那样多次调整关键词和时间范围，逐步观察结果分布。
+- 用户已经知道 index，但当前只需要基于关键词快速找事件，不需要精确字段过滤。
+
+使用方法：
+
+- 必填参数是 `keyword`、`time_range_start`、`time_range_end`。
+- `keyword` 可以是单个字符串，也可以是字符串列表；传列表时表示 AND 匹配。
+- `index_name` 可选；不传表示全局搜索，传入时表示只在指定源或 index 中搜索。
+- `time_field` 默认使用 `@timestamp`，只有在已知数据源使用其他时间字段时才改。
+
+适合输出和解读：
+
+- 返回的是搜索命中的事件集合，适合观察是否命中、命中分布在哪些 backend 或 index、事件时间是否集中、是否出现新的可疑关键词或字段。
+- 它既可以作为调查起点，也可以作为调查中的快速二次搜索工具。
+- 它不要求调用前就知道精确字段名，因此更适合模糊搜索和探索。
+
+### `siem_adaptive_query`
+
+适用场景：
+
+- 用户已经知道或基本确认日志所在的 `index_name`。
+- 用户已经有较明确的过滤条件，例如某个字段必须等于某值，或某几个字段要做精确匹配。
+- 用户想做 top-N、分组统计、字段聚合、受控范围查询。
+- 用户需要一个更稳定、可复现、结构化的查询，而不是继续做自由关键词搜索。
+
+使用方法：
+
+- 必填参数是 `index_name`、`time_range_start`、`time_range_end`。
+- `filters` 是精确字段过滤条件，键是字段名，值可以是单个字符串或字符串列表。
+- `aggregation_fields` 是可选的聚合字段列表；只有用户明确需要统计或分组时才添加。
+- `time_field` 默认使用 `@timestamp`，只有在目标数据源不是这个字段时才修改。
+
+适合输出和解读：
+
+- 返回的是结构化查询结果，更适合做稳定验证、字段统计和结论支撑。
+- 它更像提供给模型使用的一个 SIEM API，不要求一定来自 `siem_keyword_search` 的下一步，但通常需要更清晰的上下文。
+- 当用户已经给出 index、时间范围和大致条件时，可以直接使用，不必强行先走关键词搜索。
+
+### 两者如何选择
+
+- 如果线索主要是“关键词”，优先用 `siem_keyword_search`。
+- 如果线索主要是“已知 index + 已知字段条件”，优先用 `siem_adaptive_query`。
+- 如果用户目标是找事件、看分布、补线索，优先用 `siem_keyword_search`。
+- 如果用户目标是精确过滤、聚合统计、稳定复现，优先用 `siem_adaptive_query`。
+- 如果用户没有说明字段名且 index 也不确定，不要强推 `siem_adaptive_query`。
+- 如果用户已经说明明确的 index 和 filters，也不要机械地先跑 `siem_keyword_search`。
+
+## 决策流程
+
+1. 如果用户问该用哪个索引、有哪些字段，或 SIEM 源如何组织，使用 `siem_explore_schema`。
+2. 如果用户给出相对时间窗口，调用 `get_current_time`，从返回的本地时间加时区推导出可用 UTC 范围，再继续。
+
+## SOP
+
+### 探索 Schema
+
+1. 如果用户不知道目标源，先调用 `siem_explore_schema()`。
+2. 如果用户已经知道索引且想要字段结构，调用 `siem_explore_schema(target_index=<index>)`。
+3. 解析返回的 JSON。
+4. 总结与调查目标最相关的索引、时间字段候选和高信号字段。
+5. 推荐下一步查询路径：先全局关键词搜索，或在已明确位置时直接做自适应查询。
+
+### 使用 `siem_keyword_search`
+
+1. 先提取已知最强关键词。
+2. 只有当用户真正要求所有条件都匹配时，才把多个关键词规范化为 AND 集合。
+3. 要求 UTC 时间戳以 `Z` 结尾。
+4. 如果用户未指定 `index_name`，可先全局搜索；如果用户已明确数据源，也可以直接限定到该 index。
+5. 调用 `siem_keyword_search`。
+6. 解析每条返回的 JSON 字符串，并按用户目标决定输出重点：
+   - 如果用户在找事件，优先展示代表性命中。
+   - 如果用户在判断日志落点，优先展示 backend、index 分布和时间分布。
+   - 如果用户在继续收敛，优先总结下一轮应补充或移除哪些关键词。
+7. 根据结果进行下一轮判断：
+  - 命中太多时，先收窄时间范围，再补充一到两个高信号关键词。
+  - 命中太少或为空时，先移除一个限制性关键词，或适度扩展时间范围。
+  - 命中集中在少数 index 时，下一轮优先指定 `index_name`。
+  - 当结果已经足够说明日志位置和关键字段时，切换到 `siem_adaptive_query`。
+
+### 使用 `siem_adaptive_query`
+
+1. 要求 `index_name`、UTC 时间范围，以及至少一个精确过滤条件或明确聚合目标。
+2. 把过滤条件规范化为精确字段/值对。
+3. 只有当用户想要流行度、top-N 统计或分组范围时，才添加 `aggregation_fields`。
+4. 调用 `siem_adaptive_query`。
+5. 用分析师语言总结过滤范围、命中情况和任何聚合输出。
+6. 如果结果不理想，优先判断是 `filters` 太严、字段名不对、时间范围不对，还是其实应回到 `siem_keyword_search` 补充上下文。
+
+### 优化搜索
+
+首选优化动作：
+
+1. 先用 `siem_keyword_search` 在全局或较大范围内建立感觉，不要过早假设 index。
+2. 在添加很多新关键词前，先收窄时间范围。
+3. 添加一两个高信号关键词，而不是很多弱关键词。
+4. 如果查询为空，移除一个限制性关键词，或适度扩大时间窗口。
+5. 当广泛搜索返回太多无关数据时，优先指定 `index_name` 或补充更强信号。
+6. 当结果已经显示日志主要集中在某类 index 时，下一轮搜索就应该收敛到这些 index。
+7. 当用户已经学到足够字段结构和日志落点时，可切换到 `siem_adaptive_query`；但如果用户仍只是想继续搜事件，也可以继续用 `siem_keyword_search`。
+8. 持续迭代直到结果质量匹配用户目标。
+
+
+
+## 回复策略
+
+始终解释搜索的含义，而不只是它返回了什么。
+
+首选回复结构：
+
+### 搜索概览
+
+- 搜索模式：schema 探索、关键词搜索或自适应查询
+- 关键词集或精确过滤条件
+- 时间范围
+- 搜索的索引或 `all`
+- 如果使用了聚合字段
+- 用一两句话给出整体解释，说明当前搜索是在探索、收敛还是验证
+
+
+### 证据要点
+
+- 对调查重要的关键字段统计。
+- 只有在增加价值时才给出代表性记录。
+- 对于 schema 探索，只强调对 hunt 重要的索引和字段。
+- 对于关键词搜索，说明这是关键词匹配结果，适合继续搜索、看分布、找线索或定位日志来源。
+- 对于自适应查询，说明这是结构化过滤或聚合结果，适合做验证、统计和稳定复现。
+
+### 下一步最佳动作
+
+- 收窄时间范围
+- 添加一个更强关键词
+- 移除一个限制性关键词
+- 根据分布切换到特定 index
+- 搜索特定索引
+- 用精确过滤切换到自适应查询
+- 把有用 SIEM 结果保存为相关 case、alert 或 artifact 上的 enrichment
+- 停止，因为证据已经足够
+
+## 澄清规则
+
+- 如果缺少时间范围，询问时间范围。
+- 只有当用户没有提供 UTC 且意图时区不清楚时，才询问时区。
+- 除非已经明显收敛，否则不要一开始就强制用户提供 `index_name`。
+- 只有当广泛搜索可能浪费、用户已经暗示已知源，或自适应查询是正确工具时，才询问 `index_name`。
+- 只有当用户想要自适应查询且 schema 仍不清楚时，才询问精确字段名。
+- 如果用户说"看看这个事件周围"，从可用 IOC 和时间框架推导出合理的首次搜索，而不是让他们设计查询。
+
+## 输出规则
+
+- 保持简洁。
+- 默认不要倾倒每条返回记录。
+- 优先展示最相关的记录和统计。
+- 当返回多个组时，按 backend 和 index 分组结果。
+- 对于 schema 探索，呈现候选列表而不是原始字段清单。
+- 对于关键词搜索，优先输出能指导下一轮搜索的分布和线索，而不是机械罗列日志。
+- 对于自适应查询，优先输出能支撑结论的过滤命中和聚合结果。
+- 如果没有找到数据，直接说明并建议最可能有用的调整。
+
+## 失败处理
+
+- 无效时间格式：要求 UTC ISO8601 带尾部 `Z`。
+- 空结果：先扩展时间范围或移除一个关键词；如果过早指定了 index，也考虑回到全局关键词搜索。
+- 太多命中：先收窄时间范围，再添加信号，必要时切到更可能的 index。
+- 未知索引或字段选择：在猜测前使用 `siem_explore_schema`。
+- Backend 或源问题：如果结果指示了，说明哪个 backend 或索引失败。
diff --git a/PLUGINS/SIRP/sirpbasemodel.py b/PLUGINS/SIRP/sirpbasemodel.py
new file mode 100644
index 0000000..e69de29
diff --git a/PLUGINS/SIRP/sirpcoremodel.py b/PLUGINS/SIRP/sirpcoremodel.py
new file mode 100644
index 0000000..e69de29
diff --git a/PLUGINS/SIRP/sirpextramodel.py b/PLUGINS/SIRP/sirpextramodel.py
new file mode 100644
index 0000000..e69de29
diff --git a/PLUGINS/SIRP/sirpmodel.py b/PLUGINS/SIRP/sirpmodel.py
deleted file mode 100644
index 24ec54e..0000000
--- a/PLUGINS/SIRP/sirpmodel.py
+++ /dev/null
@@ -1,686 +0,0 @@
-from __future__ import annotations
-
-from datetime import datetime, timezone
-from enum import StrEnum
-from typing import List, Optional, Any, Union, ClassVar, Annotated
-
-from pydantic import field_validator, ConfigDict, BaseModel, BeforeValidator, Field, PlainSerializer
-
-from PLUGINS.SIRP.nocolymodel import AttachmentModel, AttachmentCreateModel
-
-
-class MessageType(StrEnum):
-    SYSTEM = "SystemMessage"
-    HUMAN = "HumanMessage"
-    TOOL = "ToolMessage"
-    AI = "AIMessage"
-
-
-class PlaybookType(StrEnum):
-    CASE = "CASE"
-    ALERT = "ALERT"
-    ARTIFACT = "ARTIFACT"
-
-
-class KnowledgeSource(StrEnum):
-    MANUAL = "Manual"
-    CASE = "Case"
-
-
-class TicketStatus(StrEnum):
-    UNKNOWN = 'Unknown'
-    NEW = 'New'
-    IN_PROGRESS = 'In Progress'
-    NOTIFIED = 'Notified'
-    ON_HOLD = 'On Hold'
-    RESOLVED = 'Resolved'
-    CLOSED = 'Closed'
-    CANCELED = 'Canceled'
-    REOPENED = 'Reopened'
-    OTHER = 'Other'
-
-
-class TicketType(StrEnum):
-    OTHER = 'Other'
-    JIRA = 'Jira'
-    SERVICENOW = 'ServiceNow'
-    PAGERDUTY = 'PagerDuty'
-    SLACK = 'Slack'
-
-
-class ArtifactType(StrEnum):
-    UNKNOWN = 'Unknown'
-    HOSTNAME = 'Hostname'
-    IP_ADDRESS = 'IP Address'
-    MAC_ADDRESS = 'MAC Address'
-    USER_NAME = 'User Name'
-    EMAIL_ADDRESS = 'Email Address'
-    URL_STRING = 'URL String'
-    FILE_NAME = 'File Name'
-    HASH = 'Hash'
-    PROCESS_NAME = 'Process Name'
-    RESOURCE_UID = 'Resource UID'
-    PORT = 'Port'
-    SUBNET = 'Subnet'
-    COMMAND_LINE = 'Command Line'
-    COUNTRY = 'Country'
-    PROCESS_ID = 'Process ID'
-    HTTP_USER_AGENT = 'HTTP User-Agent'
-    CWE = 'CWE'
-    CVE = 'CVE'
-    USER_CREDENTIAL_ID = 'User Credential ID'
-    ENDPOINT = 'Endpoint'
-    USER = 'User'
-    EMAIL = 'Email'
-    UNIFORM_RESOURCE_LOCATOR = 'Uniform Resource Locator'
-    FILE = 'File'
-    PROCESS = 'Process'
-    GEO_LOCATION = 'Geo Location'
-    CONTAINER = 'Container'
-    REGISTRY = 'Registry'
-    FINGERPRINT = 'Fingerprint'
-    GROUP = 'Group'
-    ACCOUNT = 'Account'
-    SCRIPT_CONTENT = 'Script Content'
-    SERIAL_NUMBER = 'Serial Number'
-    RESOURCE = 'Resource'
-    MESSAGE = 'Message'
-    ADVISORY = 'Advisory'
-    FILE_PATH = 'File Path'
-    DEVICE = 'Device'
-    REGISTRY_PATH = "Registry Path"
-    OTHER = 'Other'
-
-
-class ArtifactRole(StrEnum):
-    UNKNOWN = 'Unknown'
-    TARGET = 'Target'
-    ACTOR = 'Actor'
-    AFFECTED = 'Affected'
-    RELATED = 'Related'
-    OTHER = 'Other'
-
-
-class ArtifactReputationScore(StrEnum):
-    UNKNOWN = 'Unknown'
-    VERY_SAFE = 'Very Safe'
-    SAFE = 'Safe'
-    PROBABLY_SAFE = 'Probably Safe'
-    LEANS_SAFE = 'Leans Safe'
-    MAY_NOT_BE_SAFE = 'May not be Safe'
-    EXERCISE_CAUTION = 'Exercise Caution'
-    SUSPICIOUS_RISKY = 'Suspicious/Risky'
-    POSSIBLY_MALICIOUS = 'Possibly Malicious'
-    PROBABLY_MALICIOUS = 'Probably Malicious'
-    MALICIOUS = 'Malicious'
-    OTHER = 'Other'
-
-
-class Severity(StrEnum):
-    UNKNOWN = "Unknown"
-    INFORMATIONAL = "Informational"
-    LOW = "Low"
-    MEDIUM = "Medium"
-    HIGH = "High"
-    CRITICAL = "Critical"
-    FATAL = "Fatal"
-    OTHER = "Other"
-
-
-class AttackStage(StrEnum):
-    RECONNAISSANCE = "Reconnaissance"
-    RESOURCE_DEVELOPMENT = "Resource Development"
-    INITIAL_ACCESS = "Initial Access"
-    EXECUTION = "Execution"
-    PERSISTENCE = "Persistence"
-    PRIVILEGE_ESCALATION = "Privilege Escalation"
-    DEFENSE_EVASION = "Defense Evasion"
-    CREDENTIAL_ACCESS = "Credential Access"
-    DISCOVERY = "Discovery"
-    LATERAL_MOVEMENT = "Lateral Movement"
-    COLLECTION = "Collection"
-    COMMAND_AND_CONTROL = "Command and Control"
-    EXFILTRATION = "Exfiltration"
-    IMPACT = "Impact"
-
-
-class Impact(StrEnum):
-    UNKNOWN = "Unknown"
-    LOW = "Low"
-    MEDIUM = "Medium"
-    HIGH = "High"
-    CRITICAL = "Critical"
-    OTHER = "Other"
-
-
-class Disposition(StrEnum):
-    UNKNOWN = "Unknown"
-    ALLOWED = "Allowed"
-    BLOCKED = "Blocked"
-    QUARANTINED = "Quarantined"
-    ISOLATED = "Isolated"
-    DELETED = "Deleted"
-    DROPPED = "Dropped"
-    CUSTOM_ACTION = "Custom Action"
-    APPROVED = "Approved"
-    RESTORED = "Restored"
-    EXONERATED = "Exonerated"
-    CORRECTED = "Corrected"
-    PARTIALLY_CORRECTED = "Partially Corrected"
-    UNCORRECTED = "Uncorrected"
-    DELAYED = "Delayed"
-    DETECTED = "Detected"
-    NO_ACTION = "No Action"
-    LOGGED = "Logged"
-    TAGGED = "Tagged"
-    ALERT = "Alert"
-    COUNT = "Count"
-    RESET = "Reset"
-    CAPTCHA = "Captcha"
-    CHALLENGE = "Challenge"
-    ACCESS_REVOKED = "Access Revoked"
-    REJECTED = "Rejected"
-    UNAUTHORIZED = "Unauthorized"
-    ERROR = "Error"
-    OTHER = "Other"
-
-
-class AlertAction(StrEnum):
-    UNKNOWN = "Unknown"
-    ALLOWED = "Allowed"
-    DENIED = "Denied"
-    OBSERVED = "Observed"
-    MODIFIED = "Modified"
-    OTHER = "Other"
-
-
-class Confidence(StrEnum):
-    UNKNOWN = "Unknown"
-    LOW = "Low"
-    MEDIUM = "Medium"
-    HIGH = "High"
-    OTHER = "Other"
-
-
-class AlertAnalyticType(StrEnum):
-    UNKNOWN = "Unknown"
-    RULE = "Rule"
-    BEHAVIORAL = "Behavioral"
-    STATISTICAL = "Statistical"
-    LEARNING = "Learning (ML/DL)"
-    FINGERPRINTING = "Fingerprinting"
-    TAGGING = "Tagging"
-    KEYWORD_MATCH = "Keyword Match"
-    REGULAR_EXPRESSIONS = "Regular Expressions"
-    EXACT_DATA_MATCH = "Exact Data Match"
-    PARTIAL_DATA_MATCH = "Partial Data Match"
-    INDEXED_DATA_MATCH = "Indexed Data Match"
-    OTHER = "Other"
-
-
-class AlertAnalyticState(StrEnum):
-    UNKNOWN = "Unknown"
-    ACTIVE = "Active"
-    SUPPRESSED = "Suppressed"
-    EXPERIMENTAL = "Experimental"
-    OTHER = "Other"
-
-
-class ProductCategory(StrEnum):
-    DLP = "DLP"
-    EMAIL = "Email"
-    OT = "OT"
-    PROXY = "Proxy"
-    UEBA = "UEBA"
-    TI = "TI"
-    IAM = "IAM"
-    EDR = "EDR"
-    NDR = "NDR"
-    CLOUD = "Cloud"
-    SIEM = "SIEM"
-    WAF = "WAF"
-    OTHER = "Other"
-
-
-class AlertPolicyType(StrEnum):
-    IDENTITY_POLICY = "Identity Policy"
-    RESOURCE_POLICY = "Resource Policy"
-    SERVICE_CONTROL_POLICY = "Service Control Policy"
-    ACCESS_CONTROL_POLICY = "Access Control Policy"
-    OTHER = "Other"
-
-
-class AlertRiskLevel(StrEnum):
-    INFO = "Info"
-    LOW = "Low"
-    MEDIUM = "Medium"
-    HIGH = "High"
-    CRITICAL = "Critical"
-    OTHER = "Other"
-
-
-class AlertStatus(StrEnum):
-    UNKNOWN = "Unknown"
-    NEW = "New"
-    IN_PROGRESS = "In Progress"
-    SUPPRESSED = "Suppressed"
-    RESOLVED = "Resolved"
-    ARCHIVED = "Archived"
-    DELETED = "Deleted"
-    OTHER = "Other"
-
-
-class CasePriority(StrEnum):
-    UNKNOWN = "Unknown"
-    LOW = "Low"
-    MEDIUM = "Medium"
-    HIGH = "High"
-    CRITICAL = "Critical"
-    OTHER = "Other"
-
-
-class CaseStatus(StrEnum):
-    NEW = "New"
-    IN_PROGRESS = "In Progress"
-    ON_HOLD = "On Hold"
-    RESOLVED = "Resolved"
-    CLOSED = "Closed"
-
-
-class CaseVerdict(StrEnum):
-    UNKNOWN = "Unknown"
-    FALSE_POSITIVE = "False Positive"
-    TRUE_POSITIVE = "True Positive"
-    DISREGARD = "Disregard"
-    SUSPICIOUS = "Suspicious"
-    BENIGN = "Benign"
-    TEST = "Test"
-    INSUFFICIENT_DATA = "Insufficient Data"
-    SECURITY_RISK = "Security Risk"
-    MANAGED_EXTERNALLY = "Managed Externally"
-    DUPLICATE = "Duplicate"
-    OTHER = "Other"
-
-
-class PlaybookJobStatus(StrEnum):
-    SUCCESS = 'Success'
-    FAILED = 'Failed'
-    PENDING = 'Pending'
-    RUNNING = 'Running'
-
-
-class KnowledgeAction(StrEnum):
-    STORE = 'Store'
-    REMOVE = 'Remove'
-    DONE = 'Done'
-
-
-class AccountModel(BaseModel):
-    accountId: Optional[str] = Field(default=None, description="User's unique identifier ID (用户的唯一标识ID)")
-    avatar: Optional[str] = Field(default=None, description="User avatar URL (用户头像的URL)")
-    email: Optional[str] = Field(default=None, description="User email address (用户的电子邮件地址)")
-    fullname: Optional[str] = Field(default=None, description="User full name (用户的全名)")
-    jobNumber: Optional[str] = Field(default=None, description="User job number (用户的工号)")
-    mobilePhone: Optional[str] = Field(default=None, description="User mobile phone number (用户的手机号码)")
-    status: Optional[int] = Field(default=None, description="User status, e.g., 1 means active (用户状态, 例如: 1表示正常)")
-
-
-def validate_datetime(v: Any) -> Any:
-    if not v:
-        return None
-    if isinstance(v, datetime):
-        return v
-    if not isinstance(v, str):
-        return v
-
-    value = v.strip()
-    if not value:
-        return None
-
-    local_tz = datetime.now().astimezone().tzinfo or timezone.utc
-
-    try:
-        dt = datetime.fromisoformat(value.replace("Z", "+00:00"))
-        if dt.tzinfo is None:
-            dt = dt.replace(tzinfo=local_tz)
-        return dt
-    except ValueError:
-        pass
-
-    try:
-        return datetime.strptime(value, "%Y-%m-%d %H:%M:%S").replace(tzinfo=local_tz)
-    except ValueError:
-        raise ValueError(f"Unsupported datetime format: {value}")
-
-
-def serialize_datetime(v: Any) -> Any:
-    if isinstance(v, datetime):
-        local_tz = datetime.now().astimezone().tzinfo or timezone.utc
-        dt = v.replace(tzinfo=local_tz) if v.tzinfo is None else v
-        return dt.astimezone(timezone.utc).strftime("%Y-%m-%dT%H:%M:%SZ")
-    return v
-
-
-AutoDatetime = Annotated[
-    Union[datetime, None],
-    BeforeValidator(validate_datetime),
-    PlainSerializer(serialize_datetime, when_used="json")
-]
-
-
-def validate_account(v: Any) -> Any:
-    if not v:
-        return ""
-    if isinstance(v, dict):
-        return v.get("fullname")
-    if isinstance(v, list):
-        if len(v) == 0:
-            return ""
-        elif len(v) == 1:
-            if isinstance(v[0], str):
-                return v[0]
-            elif isinstance(v[0], dict):
-                return v[0].get("fullname")
-            else:
-                raise ValueError(f"Unsupported account format: {v}")
-        else:
-            tmp = []
-            for one in v:
-                if isinstance(one, str):
-                    tmp.append(one)
-                elif isinstance(one, dict):
-                    tmp.append(one.get("fullname"))
-                else:
-                    raise ValueError(f"Unsupported account format: {one}")
-            return tmp
-
-    if not isinstance(v, str):
-        return v
-    raise ValueError(f"Unsupported account format: {v}")
-
-
-AutoAccount = Annotated[
-    str,
-    BeforeValidator(validate_account),
-]
-
-
-class BaseSystemModel(BaseModel):
-    model_config = ConfigDict(populate_by_name=True)
-    _AI_EXCLUDE_FIELDS: ClassVar[set[str]] = {"row_owner", "row_createdBy", "row_createdAt", "row_updatedAt", "row_updatedBy"}  # 不传递给 AI 的字段
-
-    row_id: Optional[str] = Field(default=None, description="Unique row ID (唯一行 ID)")
-    row_owner: Optional[AutoAccount] = Field(default=None, description="Record owner (记录所有者)")
-    row_createdBy: Optional[AutoAccount] = Field(default=None, description="Creator (创建者)")
-    row_createdAt: Optional[AutoDatetime] = Field(default=None, description="Record created time (记录创建时间)")
-    row_updatedAt: Optional[AutoDatetime] = Field(default=None, description="Record last updated time (记录最后更新时间)")
-    row_updatedBy: Optional[AutoAccount] = Field(default=None, description="Last updated by (最后更新人)")
-
-    def __init_subclass__(cls, **kwargs):
-        super().__init_subclass__(**kwargs)
-        # 将当前类的集合与父类的集合自动合并
-        cls._AI_EXCLUDE_FIELDS = cls._AI_EXCLUDE_FIELDS | BaseSystemModel._AI_EXCLUDE_FIELDS
-
-    @field_validator("row_owner", mode="before")
-    @classmethod
-    def empty_list_to_none(cls, v: Any) -> Any:
-        if isinstance(v, list) and len(v) == 0:
-            return None
-        return v
-
-    def model_dump_for_ai(
-            self,
-            *,
-            exclude_none: bool = True,
-            exclude_unset: bool = True,
-            exclude_default: bool = True,
-    ) -> dict[str, Any]:
-        return self.model_dump(
-            exclude=self._AI_EXCLUDE_FIELDS,
-            exclude_none=exclude_none,
-            exclude_unset=exclude_unset,
-            exclude_defaults=exclude_default,
-            by_alias=True
-        )
-
-    def model_dump_json_for_ai(
-            self,
-            *,
-            exclude_none: bool = True,
-            exclude_unset: bool = True,
-            exclude_default: bool = True,
-    ) -> str:
-        return self.model_dump_json(
-            exclude=self._AI_EXCLUDE_FIELDS,
-            exclude_none=exclude_none,
-            exclude_unset=exclude_unset,
-            exclude_defaults=exclude_default,
-            by_alias=True
-        )
-
-
-class MessageModel(BaseSystemModel):
-    playbook: Optional[List[Union[PlaybookModel, str]]] = Field(default="", description="Owning playbook row ID (所属剧本行 ID)")
-    node: Optional[str] = Field(default="", description="Source node name or ID (源节点名称或 ID)")
-    content: Optional[str] = Field(default="", description="Message text content (消息文本内容)")
-    data: Optional[str] = Field(default="", description="Message JSON payload (消息 JSON 负载)")
-    type: Optional[MessageType] = Field(default=None, description="Message role type (消息角色类型)")
-
-
-class PlaybookModel(BaseSystemModel):
-    id: Optional[str] = Field(default=None, description="Record ID e.g. playbook_000001 (记录 ID e.g. playbook_000001)")
-    source_row_id: Optional[str] = Field(default="", description="Trigger source row ID (触发源行 ID)")
-    source_id: Optional[str] = Field(default="",
-                                     description="Trigger source record ID e.g. case_00000_1,alert_000001,artifact_000001 (触发源记录 ID e.g. case_00000_1,alert_000001,artifact_000001)")
-    type: Optional[PlaybookType] = Field(default=None, description="Linked object type (关联对象类型)")
-    name: Optional[str] = Field(default="", description="Executed playbook name (执行剧本名称)")
-    user_input: Optional[str] = Field(default="", description="Initial or follow-up user input (初始或后续用户输入)")
-    user: Optional[AutoAccount] = Field(default=None, description="Playbook requester (剧本请求者)")
-
-    job_status: Optional[PlaybookJobStatus] = Field(default=None, description="Background job status (后台任务状态)")
-    job_id: Optional[str] = Field(default="", description="Background job ID (后台任务 ID)")
-    remark: Optional[str] = Field(default="", description="Execution remark (执行备注)")
-
-    # 关联表
-    messages: Optional[List[Union[MessageModel, str]]] = Field(default=None, description="Execution message history (执行消息历史)")
-
-
-class KnowledgeModel(BaseSystemModel):
-    id: Optional[str] = Field(default=None, description="Record ID e.g. knowledge_000001 (记录 ID e.g. knowledge_000001)")
-    title: Optional[str] = Field(default="", description="Knowledge title (知识标题)")
-    body: Optional[str] = Field(default="", description="Knowledge content (知识内容)")
-    using: Optional[bool] = Field(default=False, description="Currently in use (当前正在使用)")
-    action: Optional[KnowledgeAction] = Field(default=None, description="Knowledge action (知识操作)")
-    source: Optional[KnowledgeSource] = Field(default=None, description="Knowledge source (知识来源)")
-    tags: Optional[List[str]] = Field(default=[], description="Knowledge tags (知识标签)", json_schema_extra={"type": 2})
-
-
-class EnrichmentModel(BaseSystemModel):
-    _AI_EXCLUDE_FIELDS: ClassVar[set[str]] = set()
-    id: Optional[str] = Field(default=None, description="Record ID e.g. enrichment_000001 (记录 ID e.g. enrichment_000001)")
-    name: Optional[str] = Field(default="", description="Enrichment name (富化名称)")
-    type: Optional[str] = Field(default="Other", description="Enrichment type (富化类型)", json_schema_extra={"type": 2})
-    provider: Optional[str] = Field(default="Other", description="Enrichment provider (富化提供商)", json_schema_extra={"type": 2})
-    value: Optional[str] = Field(default="", description="Enrichment value (富化值)")
-    src_url: Optional[str] = Field(default="", description="Enrichment source URL (富化来源 URL)")
-    desc: Optional[str] = Field(default="", description="Enrichment summary (富化摘要)")
-    data: Optional[str] = Field(default="", description="Detailed enrichment JSON (详细富化 JSON)")
-
-
-class TicketModel(BaseSystemModel):
-    _AI_EXCLUDE_FIELDS: ClassVar[set[str]] = {'case'}
-
-    id: Optional[str] = Field(default=None, description="Record ID e.g. ticket_000001 (记录 ID e.g. ticket_000001)")
-    status: Optional[TicketStatus] = Field(
-        default=None, description="External ticket status (外部工单状态)")
-    type: Optional[TicketType] = Field(default=None, description="External ticket type (外部工单类型)",
-                                       json_schema_extra={"type": 2})
-    title: Optional[str] = Field(default="", description="Ticket title (工单标题)")
-    uid: Optional[str] = Field(default="", description="External ticket ID (外部工单 ID)")
-    src_url: Optional[str] = Field(default="", description="External ticket URL (外部工单 URL)")
-
-    # 反向关联,无需手动处理
-    case: Optional[List[Union[CaseModel, str]]] = Field(default=None, description="Linked case row_id (关联案例行 ID)")
-
-
-class ArtifactModel(BaseSystemModel):
-    """Stores entity information extracted from alerts, the minimum investigatable unit"""
-    _AI_EXCLUDE_FIELDS: ClassVar[set[str]] = {'alert'}
-
-    id: Optional[str] = Field(default=None, description="Record ID e.g. artifact_000001 (记录 ID e.g. artifact_000001)")
-    name: Optional[str] = Field(default="", description="Artifact name (实体名称)")
-    type: Optional[ArtifactType] = Field(default=None, description="Artifact type (实体类型)")
-    role: Optional[ArtifactRole] = Field(default=None, description="Artifact role in event (实体在事件中的角色)")
-    value: Optional[str] = Field(default="", description="Artifact value (实体值)")
-
-    owner: Optional[str] = Field(default="", description="Owning system or user (所属系统或用户)")
-    reputation_provider: Optional[str] = Field(default="", description="Threat intel provider (威胁情报提供商)", json_schema_extra={"type": 2})
-    reputation_score: Optional[ArtifactReputationScore] = Field(default=None, description="Artifact reputation (实体信誉)")
-
-    # 反向关联,无需手动处理
-    alert: Optional[List[Union[AlertModel, str]]] = Field(default=None, description="Linked alert row_id (关联告警行 ID)")
-
-    # 关联表
-    enrichments: Optional[List[Union[EnrichmentModel, str]]] = Field(default=None,
-                                                                     description="Enrichments information (富化信息)")  # None 时表示无需处理,[] 时表示要将 link 清空
-
-
-class AlertModel(BaseSystemModel):
-    _AI_EXCLUDE_FIELDS: ClassVar[set[str]] = {'raw_data', 'severity_ai', 'confidence_ai', "impact_ai", "comment_ai", "case"}
-    # 系统自动生成字段
-    id: Optional[str] = Field(default=None,
-                              description="Record ID e.g. alert_000001, auto-generated, no manual input needed (记录 ID e.g. alert_000001, 系统自动生成,无需手动赋值)")
-
-    # 创建记录填写字段
-    title: Optional[str] = Field(default="", description="Alert title (告警标题)")
-    severity: Optional[Severity] = Field(default=Severity.UNKNOWN, description="Source-defined severity (告警来源定义的严重程度)")
-    confidence: Optional[Confidence] = Field(default=Confidence.UNKNOWN, description="True-positive confidence (真阳性置信度)")
-    impact: Optional[Impact] = Field(default=Impact.UNKNOWN, description="Potential impact (告警潜在影响)")
-    disposition: Optional[Disposition] = Field(default=Disposition.UNKNOWN, description="Source disposition (告警源处置结果)")
-    action: Optional[AlertAction] = Field(default=AlertAction.UNKNOWN, description="Observed action (告警源的动作)")
-
-    labels: Optional[List[str]] = Field(default=[], description="Alert labels (告警标签)", json_schema_extra={"type": 2})
-    desc: Optional[str] = Field(default="", description="Alert description (告警描述)")
-
-    first_seen_time: Optional[AutoDatetime] = Field(default=None, description="First observed time (首次观测时间)")
-    last_seen_time: Optional[AutoDatetime] = Field(default=None, description="Last observed time (最后观测时间)")
-
-    rule_id: Optional[str] = Field(default="", description="SIEM rule ID (SIEM 规则 ID)")
-    rule_name: Optional[str] = Field(default="", description="SIEM rule name (SIEM 规则名称)")
-    correlation_uid: Optional[str] = Field(default="",
-                                           description="Case correlation ID, alerts with the same correlation_uid are linked to the same event (事件关联 ID,相同 correlation_uid 告警关联到同一个事件)")
-
-    src_url: Optional[str] = Field(default="", description="Source alert URL (原始告警 URL)")
-    source_uid: Optional[str] = Field(default="",
-                                      description="Source product ID, can be used to locate the unique alert in the source system (原始告警 唯一ID, 可通过该 ID 在原始来源中定位唯一告警)")
-    data_sources: Optional[List[str]] = Field(default=[], description="Underlying data sources (告警源生成告警的数据来源列表)")
-
-    analytic_name: Optional[str] = Field(default="", description="Analytic engine name (分析引擎名称)")
-    analytic_type: Optional[AlertAnalyticType] = Field(default=AlertAnalyticType.UNKNOWN, description="Analytic engine type (分析引擎类型)")
-    analytic_state: Optional[AlertAnalyticState] = Field(default=None, description="Analytic rule state (分析规则状态)")
-    analytic_desc: Optional[str] = Field(default="", description="Analytic rule description (分析规则描述)")
-
-    tactic: Optional[str] = Field(default="", description="Mapped MITRE tactic (映射的 MITRE 战术)")
-    technique: Optional[str] = Field(default="", description="Mapped MITRE technique (映射的 MITRE 技术)")
-    sub_technique: Optional[str] = Field(default="", description="Mapped MITRE sub-technique (映射的 MITRE 子技术)")
-    mitigation: Optional[str] = Field(default="", description="Suggested mitigation (建议的缓解措施)")
-
-    product_category: Optional[ProductCategory] = Field(default=None, description="Source product category (原始产品类别)")
-    product_vendor: Optional[str] = Field(default=None, description="Source vendor (原始厂商)", json_schema_extra={"type": 2})
-    product_name: Optional[str] = Field(default=None, description="Source product name (原始产品名称)", json_schema_extra={"type": 2})
-    product_feature: Optional[str] = Field(default=None, description="Source product feature (原始产品功能)", json_schema_extra={"type": 2})
-
-    policy_name: Optional[str] = Field(default="", description="Trigger policy name (触发策略名称)")
-    policy_type: Optional[AlertPolicyType] = Field(default=None, description="Trigger policy type (触发策略类型)")
-    policy_desc: Optional[str] = Field(default="", description="Trigger policy description (触发策略描述)")
-
-    risk_level: Optional[AlertRiskLevel] = Field(default=None, description="Assessed risk level (评估的风险等级)")
-
-    status: Optional[AlertStatus] = Field(default=None, description="Alert handling status (告警处理状态)")
-    status_detail: Optional[str] = Field(default="", description="Handling status details (处理状态详情)")
-    remediation: Optional[str] = Field(default="", description="Remediation advice or record (处置建议或记录)")
-
-    unmapped: Optional[str] = Field(default="", description="Raw unmapped fields, JSON Format (原始未映射字段 JSON 格式)")
-
-    raw_data: Optional[str] = Field(default="", description="Raw alert log JSON (原始告警日志 JSON)")
-
-    # AI字段
-    severity_ai: Optional[Severity] = Field(default=None, description="AI-assessed severity (AI 评估严重程度)")
-    confidence_ai: Optional[Confidence] = Field(default=None, description="AI-assessed confidence (AI 评估置信度)")
-    impact_ai: Optional[Impact] = Field(default=Impact.UNKNOWN, description="AI-assessed potential impact (AI 评估潜在影响)")
-    risk_level_ai: Optional[AlertRiskLevel] = Field(default=None, description="Assessed risk level (AI 评估的风险等级)")
-    comment_ai: Optional[str] = Field(default="", description="AI-generated comment (AI 生成的注释)")
-
-    # 反向关联,无需手动处理
-    case: Optional[List[Union[CaseModel, str]]] = Field(default=None,
-                                                        description="Linked case row_id, reverse association, auto-linked, no manual setting needed (关联案例行 ID,反向关联,自动化关联,无需手动设置)")
-
-    artifacts: Optional[List[Union[ArtifactModel, str]]] = Field(default=None, description="Extracted artifacts (关联表, 提取的实体列表)")
-    enrichments: Optional[List[Union[AlertModel, str]]] = Field(default=None, description="Alert enrichments (关联表, 告警富化)")
-
-
-class CaseModel(BaseSystemModel):
-    _AI_EXCLUDE_FIELDS: ClassVar[set[str]] = {"attachments", "workbook", "summary_ai", "comment_ai", "attack_stage_ai", "severity_ai", "confidence_ai",
-                                              "threat_hunting_report_ai", "verdict_ai"}
-
-    id: Optional[str] = Field(default=None, description="Record ID e.g. case_000001 (记录 ID e.g. case_000001)")
-    title: Optional[str] = Field(default="", description="Case title (案例标题)")
-    severity: Optional[Severity] = Field(default=None,
-                                         description="Analyst-assessed severity (分析师评估严重程度)")
-    impact: Optional[Impact] = Field(default=None, description="Analyst-assessed impact (分析师评估影响)")
-    priority: Optional[CasePriority] = Field(default=None, description="Response priority (响应优先级)")
-
-    confidence: Optional[Confidence] = Field(default=None, description="Analyst-assessed confidence (分析师评估置信度)")
-    description: Optional[str] = Field(default="", description="Case description (案例描述)")
-
-    category: Optional[ProductCategory] = Field(default=None,
-                                                description="Case category (案例类别)")
-    tags: Optional[List[str]] = Field(default=[], description="Case tags (案例标签)", json_schema_extra={"type": 2})
-
-    status: Optional[CaseStatus] = Field(default=None,
-                                         description="Case handling status (案例处理状态)")
-    assignee_l1: Optional[AutoAccount] = Field(default=None, description="Assigned L1 analyst (分配的 L1 分析师)")
-    acknowledged_time: Optional[AutoDatetime] = Field(default=None, description="L1 first acknowledged time (L1 首次接手时间)")
-    comment: Optional[str] = Field(default="", description="Case analyst comment (案例分析师注释)")
-    attachments: Optional[List[Union[AttachmentModel, AttachmentCreateModel]]] = Field(default=[], description="Case attachments (分析师上传的案例附件)")
-
-    assignee_l2: Optional[AutoAccount] = Field(default=None,
-                                               description="Assigned or escalated L2 analyst (分配或升级的 L2 分析师)")
-    assignee_l3: Optional[AutoAccount] = Field(default=None,
-                                               description="Assigned or escalated L3 analyst (分配或升级的 L3 分析师)")
-    closed_time: Optional[AutoDatetime] = Field(default=None, description="Case closed time (案例关闭时间)")
-    verdict: Optional[CaseVerdict] = Field(
-        default=None, description="Final verdict (最终判定结果)")
-    summary: Optional[str] = Field(default="", description="Closure summary (结案摘要)")
-
-    correlation_uid: Optional[str] = Field(default="", description="Case correlation ID (案例关联 ID)")
-
-    workbook: Optional[str] = Field(default="", description="Investigation workbook (调查工作手册)")
-
-    # ai 字段
-    severity_ai: Optional[Severity] = Field(default=None, description="AI-assessed severity (AI 评估严重程度)")
-    confidence_ai: Optional[Confidence] = Field(default=None, description="AI-assessed confidence (AI 评估置信度)")
-    impact_ai: Optional[Impact] = Field(default=None, description="AI-assessed impact (AI 评估影响)")
-    priority_ai: Optional[CasePriority] = Field(default=None, description="AI-assessed response priority (AI 评估响应优先级)")
-    attack_stage_ai: Optional[AttackStage] = Field(default="", description="AI-assessed attack stage (AI 评估攻击阶段)")
-    comment_ai: Optional[str] = Field(default="", description="AI-generated comment (AI 生成的注释)")
-    verdict_ai: Optional[CaseVerdict] = Field(default=None, description="AI-generated final verdict (AI 生成的最终判定结果)")
-    summary_ai: Optional[str] = Field(default="", description="AI-generated closure summary (AI 生成的结案摘要)")
-
-    # 公式自动计算字段,无需手动赋值
-    start_time_calc: Optional[Any] = Field(default=None, description="Calculated start time (计算的开始时间)")
-    end_time_calc: Optional[Any] = Field(default=None, description="Calculated end time (计算的结束时间)")
-    detect_time_calc: Optional[Any] = Field(default=None, description="Calculated detect time (计算的检测时间)")
-    acknowledge_time_calc: Optional[Any] = Field(default=None, description="Calculated acknowledge time (计算的接手时间)")
-    respond_time_calc: Optional[Any] = Field(default=None, description="Calculated response time (计算的响应时间)")
-
-    # 关联表
-    tickets: Optional[List[Union[TicketModel, str]]] = Field(default=None, description="Linked external tickets (关联外部工单)")
-    enrichments: Optional[List[Union[EnrichmentModel, str]]] = Field(default=None, description="Case enrichments (案例富化)")
-    alerts: Optional[List[Union[AlertModel, str]]] = Field(default=None, description="Merged alerts (合并的告警)")
-
-    @field_validator('attachments', mode='before')
-    def handle_attachments(cls, v):
-        if v == "":
-            return []
-        return v