windows-protocols

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Windows Protocols Corpus Navigator

Windows协议语料库导航工具

Overview

概述

The markdown corpus is already extracted locally, next to
```
SKILL.md
```
.
Use local files only. Do not use network lookup for this skill.
Do not run download or setup commands as part of normal usage.

This skill is optimized for structure-first navigation of Microsoft Open Specifications markdown and evidence-grounded answers.

Markdown语料库已本地提取，位于
```
SKILL.md
```
旁边。
仅使用本地文件。使用此Skill时不要进行网络查询。
正常使用时不要运行下载或设置命令。

此Skill针对Microsoft开放规范Markdown的结构化优先导航以及基于证据的回答进行了优化。

Corpus Layout

语料库布局

```
README.md
```
— index with three sections: Overview Documents (domain maps), Technical Documents (full spec list), and Reference Documents. Overview Documents are at the top; use them for topical discovery.
```
LEGAL.md
```
— legal and redistribution notice.
```
<PROTOCOL-ID>/
```
— protocol directories (Overview, Technical, or Reference), directly under this folder.
```
<PROTOCOL-ID>/<PROTOCOL-ID>.md
```
— primary markdown spec content.
```
<PROTOCOL-ID>/media/
```
— extracted figures and image assets referenced by the markdown.

```
README.md
```
— 索引，包含三个部分：概述文档（领域映射）、技术文档（完整规范列表）和参考文档。概述文档位于顶部；可用于主题式探索。
```
LEGAL.md
```
— 法律和再分发声明。
```
<PROTOCOL-ID>/
```
— 协议目录（概述、技术或参考类型），直接位于此文件夹下。
```
<PROTOCOL-ID>/<PROTOCOL-ID>.md
```
— 主要Markdown规范内容。
```
<PROTOCOL-ID>/media/
```
— Markdown引用的提取后的图表和图像资源。

Corpus Document Types

语料库文档类型

Overview Documents — Domain maps that group related protocols under a topic (file access, authentication, storage, remote desktop, etc.). Each describes system capabilities, lists member protocols with their roles, and contains many links to specifications. Use as the primary entry point for topical discovery.
Technical Documents — Individual protocol specifications with full wire-format, message syntax, and behavior details.
Reference Documents — Supplemental references (error codes, shared types, etc.).

概述文档 — 领域映射文档，按主题（文件访问、身份验证、存储、远程桌面等）对相关协议进行分组。每个文档都会描述系统功能、列出成员协议及其角色，并包含许多指向规范的链接。可作为主题式探索的主要入口点。
技术文档 — 单个协议规范，包含完整的有线格式、消息语法和行为细节。
参考文档 — 补充参考资料（错误代码、共享类型等）。

Protocol Clusters by Domain

按领域划分的协议集群

When using Overview documents for discovery, expect these cluster themes:

Authentication: Kerberos, NTLM, SPNEGO — core for file access, RDP, Group Policy.
File Access: SMB/CIFS family, DFS, WebDAV, shared types (FSCC, FSA).
Remote Desktop: Base + graphics pipeline, gateway, virtual channels (clipboard, devices, transport).
Storage: Disk/volume management, shadow copies, EFS, backup, removable media.
Group Policy: Core protocol + extensions; depends on AD and file access.
Network Access Protection: SoH/SoHR over PEAP; DHCP/VPN/TSGU/IPsec enforcement.
Content Caching: Peer discovery, retrieval, hosted cache; integrates with SMB/HTTP.
File Services Management: FSRM, remote VSS, classification.
Rights Management: Publish/consume protected content.
Virtual Storage: Remote shared virtual disks over SMB.

使用概述文档进行探索时，会遇到以下集群主题：

身份验证：Kerberos、NTLM、SPNEGO — 是文件访问、RDP、组策略的核心协议。
文件访问：SMB/CIFS系列、DFS、WebDAV、共享类型（FSCC、FSA）。
远程桌面：基础+图形管道、网关、虚拟通道（剪贴板、设备、传输）。
存储：磁盘/卷管理、卷影副本、EFS、备份、可移动媒体。
组策略：核心协议+扩展；依赖于AD和文件访问。
网络访问保护：基于PEAP的SoH/SoHR；DHCP/VPN/TSGU/IPsec强制实施。
内容缓存：对等发现、检索、托管缓存；与SMB/HTTP集成。
文件服务管理：FSRM、远程VSS、分类。
权限管理：发布/使用受保护内容。
虚拟存储：通过SMB实现的远程共享虚拟磁盘。

Protocol Naming and Scope

协议命名与范围

```
MS-...
```
documents are Microsoft protocol specifications (wire formats, behavior, state machines, structures).
Families share prefixes (for example
```
MS-RDP*
```
,
```
MS-AD*
```
,
```
MS-MQ*
```
).

When a user asks a broad question, start from family prefixes and narrow to specific protocol IDs.

```
MS-...
```
文档是Microsoft协议规范（有线格式、行为、状态机、结构）。
同系列协议共享前缀（例如
```
MS-RDP*
```
、
```
MS-AD*
```
、
```
MS-MQ*
```
）。

当用户提出宽泛问题时，先从系列前缀入手，再缩小到具体的协议ID。

When to Use / When Not to Use

使用场景与禁用场景

Use this skill when the user asks for:

Protocol semantics, message flow, field definitions, or sequencing rules.
Cross-reference analysis between related specs.
Security semantics and product-behavior distinctions in Open Specifications docs.

Do not use this skill as the primary source for:

API usage tutorials, SDK how-to steps, or implementation quickstarts.
Product configuration guidance not grounded in protocol specs.

If a required protocol is missing from root-level

MS-*

directories, state that explicitly and ask the user for the exact protocol ID or a narrower feature scope.

在以下场景使用此Skill：

协议语义、消息流、字段定义或排序规则相关问题。
相关规范间的交叉引用分析。
开放规范文档中的安全语义和产品行为差异。

请勿将此Skill作为以下内容的主要来源：

API使用教程、SDK操作步骤或实现快速入门。
未基于协议规范的产品配置指南。

如果根级

MS-*

目录中缺少所需的协议，请明确告知用户，并请求用户提供确切的协议ID或更窄的功能范围。

Navigation Strategy

导航策略

Domain/topic questions (file access, authentication, storage, remote desktop, etc.): Start with the corresponding Overview document from README. Use it as a map to identify which specifications to read. Follow its links into those specs.
Protocol-ID questions: Go directly to the spec if the ID is known.
Principle: Overview documents are the primary entry point for topical discovery. README provides the index of Overview docs and the full protocol list, but does not contain the semantic mapping (topic → protocols) found in Overviews.
Anti-pattern: Do not rely on README keyword search alone. A shallow grep of README may return protocol IDs from unrelated domains; overview documents provide the curated topic-to-protocol mapping.

领域/主题问题（文件访问、身份验证、存储、远程桌面等）：从README中对应的概述文档开始。将其作为映射工具来确定需要阅读的规范。跟随其中的链接进入这些规范。
协议ID相关问题：如果已知协议ID，直接进入对应的规范。
原则：概述文档是主题式探索的主要入口点。README提供了概述文档的索引和完整协议列表，但不包含概述文档中提供的语义映射（主题→协议）。
反模式：不要仅依赖README的关键词搜索。对README进行浅层 grep 可能会返回来自无关领域的协议ID；概述文档提供了经过整理的主题与协议的映射关系。

Navigation Workflow

导航工作流

For topical questions: Identify the domain from the question, match to an Overview document in README, read it to obtain linked member specs, then follow links into those specs.
For known protocol IDs: Validate in
```
README.md
```
and directory names (
```
<PROTOCOL-ID>/
```
), then open the spec.
Start from the spec TOC (
```
<summary>
```
blocks and numbered entries) before deep reading.
Read sections in this order by intent: orientation/versioning, message or structure syntax, behavior and sequencing, security and product behavior notes.
Cross-check base vs extension specs when requirements may be split.
Answer with explicit protocol IDs and exact section headings; separate confirmed facts from inference.

对于主题式问题：从问题中识别领域，匹配README中的概述文档，阅读文档以获取链接的成员规范，然后跟随链接进入这些规范。
对于已知协议ID的问题：在
```
README.md
```
和目录名称（
```
<PROTOCOL-ID>/
```
）中进行验证，然后打开对应的规范。
在深入阅读之前，先查看规范的目录（
```
<summary>
```
块和编号条目）。
按以下意图顺序阅读章节：定位/版本控制、消息或结构语法、行为和排序、安全和产品行为说明。
当需求可能分散在多个规范中时，交叉检查基础规范与扩展规范。
回答时需包含明确的协议ID和确切的章节标题；将已确认的事实与推断内容分开。

Link-Following

链接跟随规则

Overview documents contain References sections (often 1.3) and extensive inline links to member specifications. Follow these links rather than guessing paths.
The link graph (Overview → spec → related spec) is the intended navigation structure.
When a spec references another spec for types, extensions, or dependencies, follow that link to reach the authoritative source.
Continue following links until the relevant information is found; do not stop at the first document that mentions the topic.

概述文档包含参考部分（通常是1.3节）和大量指向成员规范的内联链接。请跟随这些链接，而非猜测路径。
链接图（概述→规范→相关规范）是预期的导航结构。
当某个规范引用另一个规范的类型、扩展或依赖项时，请跟随该链接以获取权威来源。
继续跟随链接直到找到相关信息；不要在第一个提到该主题的文档处停止。

Canonical Spec Structure (Corpus-Guided)

规范的标准结构（基于语料库）

The corpus is highly consistent. The most common top-level sequence is:

```
1 Introduction
```
```
2 Messages
```
```
3 Protocol Details
```
```
4 Protocol Examples
```
```
5 Security
```

Common

1.x

orientation subsections include:

```
1.1 Glossary
```
,
```
1.2 References
```
,
```
1.3 Overview
```
```
1.4 Relationship to Other Protocols
```
```
1.5 Prerequisites/Preconditions
```

1.7 Versioning and Capability Negotiation

Common

3.x.y

behavioral subsections include:

```
Abstract Data Model
```
```
Timers
```
```
Initialization
```
```
Higher-Layer Triggered Events
```

Message Processing Events and Sequencing Rules

```
Timer Events
```
```
Other Local Events
```

Use these sections as the default reading spine unless the document is an outlier.

语料库的结构高度一致。最常见的顶级章节顺序为：

```
1 引言
```
```
2 消息
```
```
3 协议细节
```
```
4 协议示例
```
```
5 安全
```

常见的

1.x

定位子章节包括：

```
1.1 术语表
```
、
```
1.2 参考资料
```
、
```
1.3 概述
```
```
1.4 与其他协议的关系
```
```
1.5 先决条件/前置条件
```
```
1.7 版本控制与能力协商
```

常见的

3.x.y

行为子章节包括：

```
抽象数据模型
```
```
计时器
```
```
初始化
```
```
高层触发事件
```
```
消息处理事件与排序规则
```
```
计时器事件
```
```
其他本地事件
```

除非文档是例外情况，否则请将这些章节作为默认的阅读主线。

Section-First Reading Playbook

章节优先阅读指南

Prioritize sections by question type:

“What is this protocol / where does it fit?” ->
```
1.3
```
,
```
1.4
```
,
```
1.7
```
“How is data encoded / what are fields?” ->
```
2.2
```
(
```
Message Syntax
```
,
```
Common Data Types
```
, structures)
“What is the runtime behavior / state machine?” ->
```
3.x
```
details and
```
3.x.y
```
sequencing sections
“What security guarantees or knobs exist?” ->
```
5.*
```
security sections
“Implementation differences by product/version?” ->
```
Appendix ... Product Behavior
```
and change tracking sections

When structure spans multiple specs, resolve in this order:

Base/core protocol semantics.
Extension-specific modifications.
Shared type dependencies (for example
```
MS-DTYP
```
).

根据问题类型优先阅读对应章节：

“此协议是什么 / 它的定位是什么？” ->
```
1.3
```
、
```
1.4
```
、
```
1.7
```
“数据如何编码 / 字段有哪些？” ->
```
2.2
```
（
```
消息语法
```
、
```
通用数据类型
```
、结构）
“运行时行为 / 状态机是怎样的？” ->
```
3.x
```
细节和
```
3.x.y
```
排序章节
“有哪些安全保障或配置选项？” ->
```
5.*
```
安全章节
“不同产品/版本的实现差异是什么？” ->
```
附录...产品行为
```
和变更跟踪章节

当结构跨越多个规范时，按以下顺序解析：

基础/核心协议语义。
扩展特定的修改。
共享类型依赖项（例如
```
MS-DTYP
```
）。

Outlier Handling

例外情况处理

Not all specs use the exact canonical headings. Handle these common variants:

2 Structures

2 Attributes

2 Data Types

2 Message Transport

instead of

2 Messages

```
3 Structure Examples
```
,
```
3 Details
```
, or algorithm-specific section names.
```
4 Security
```
or
```
4 Security Considerations
```
when numbering shifts.

Appendix variants such as

Full IDL

Full WSDL

Full XML Schema

Full JSON Schema

Change tracking may appear as section
```
7
```
,
```
8
```
,
```
9
```
, or higher.

Rule: match by semantic intent first, section number second.

并非所有规范都使用完全标准的标题。请处理以下常见变体：

使用

2 结构

、

2 属性

、

2 数据类型

、

2 消息传输

而非

2 消息

。

使用
```
3 结构示例
```
、
```
3 细节
```
或特定算法的章节名称。
当编号顺序变化时，使用
```
4 安全
```
或
```
4 安全注意事项
```
。

附录变体如

完整IDL

、

完整WSDL

、

完整XML Schema

、

完整JSON Schema

。

变更跟踪可能出现在第7、8、9或更高章节。

规则：优先匹配语义意图，其次是章节编号。

Quick Discovery Patterns

快速探索模式

When the protocol ID is unknown: Identify the domain first from question keywords, then open the matching Overview document before picking specs. Use its Protocol Summary or References section to obtain the correct spec list.
Do not select specs from README's flat Technical Documents list without consulting an Overview when the domain is ambiguous.
When the user provides a specific protocol ID: Go directly to that spec.
For extension behavior: Verify whether requirements are in a base protocol or an extension; Overview docs and spec cross-references help clarify dependencies.
For ambiguous acronyms: List 2–4 likely protocols (via Overview docs) and ask the user to choose before deep analysis.

当未知协议ID时：首先从问题关键词中识别领域，然后打开匹配的概述文档，再选择规范。使用其协议摘要或参考部分获取正确的规范列表。
当领域不明确时，不要直接从README的扁平技术文档列表中选择规范，而应先查阅概述文档。
当用户提供特定协议ID时：直接进入该规范。
对于扩展行为：验证需求是在基础协议还是扩展协议中；概述文档和规范的交叉引用有助于明确依赖关系。
对于模糊的首字母缩写：列出2-4个可能的协议（通过概述文档），并请用户选择后再进行深入分析。

Answer Contract (Semi-Structured)

回答规范（半结构化）

Default answer shape:

```
Protocols consulted
```
: list exact IDs.
```
Sections used
```
: list exact section titles (and numbers when available).
```
Findings
```
: concise facts tied to those sections.
```
Inference / uncertainty
```
: explicit separation from confirmed text.

Evidence policy:

For non-obvious, contested, or security-sensitive claims, include section-grounded evidence.
For straightforward facts, concise section references are enough.
If two specs disagree, report both and identify likely version/context scope.

默认回答格式：

```
参考的协议
```
：列出确切的ID。
```
使用的章节
```
：列出确切的章节标题（如有编号请同时列出）。
```
发现内容
```
：与这些章节相关的简洁事实。
```
推断 / 不确定性
```
：明确与已确认文本区分开。

证据策略：

对于非显而易见、有争议或安全敏感的声明，需包含基于章节的证据。
对于简单事实，简洁的章节引用即可。
如果两个规范存在分歧，请同时报告两者，并指出可能的版本/上下文范围。

Working Style

工作风格

Prefer section-grounded answers over high-level paraphrase.
Include the exact protocol IDs used for the answer.
If the corpus does not clearly answer a point, state uncertainty explicitly.
Never present inferred behavior as normative requirement text.

优先选择基于章节的回答，而非高层级的转述。
回答中需包含用于获取答案的确切协议ID。
如果语料库未明确回答某个问题，请明确说明不确定性。
切勿将推断的行为作为规范性要求文本呈现。