chronology

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

/chronology

/时间线构建

Load

~/.claude/plugins/config/claude-for-legal/litigation-legal/matters/[slug]/matter.md

→ theory, pivot fact, key facts.

Load

~/.claude/plugins/config/claude-for-legal/litigation-legal/CLAUDE.md

→ Document storage sources, default matter folder pattern.

Follow the workflow and reference below.
Identify sources in order: user-provided paths this session, default matter folder, declared sources from
```
~/.claude/plugins/config/claude-for-legal/litigation-legal/CLAUDE.md
```
.
For readable sources: extract dated events. For unreachable sources: note in Gaps.
De-dupe, merge with sources list per event.
Tag significance (🔴/🟡/⚪) per matter theory.

Write

~/.claude/plugins/config/claude-for-legal/litigation-legal/matters/[slug]/chronology.md

(or format variant per flag).

If prior version exists: version number increments, diff summary presented to user.
Confirm before finalizing: "Here's what I built. Scan the 🔴 entries — anything I miscalled?"

加载

~/.claude/plugins/config/claude-for-legal/litigation-legal/matters/[slug]/matter.md

→ 获取案件理论、核心事实、关键事实。

加载

~/.claude/plugins/config/claude-for-legal/litigation-legal/CLAUDE.md

→ 获取文档存储源、默认案件文件夹规则。

遵循下方工作流程和参考说明。
按以下顺序识别源：本次会话中用户提供的路径、默认案件文件夹、
```
~/.claude/plugins/config/claude-for-legal/litigation-legal/CLAUDE.md
```
中指定的源。
对于可读取的源：提取带日期的事件。对于无法访问的源：在「缺口」部分标注。
去重，并按事件合并源列表。
根据案件理论标记重要性（🔴/🟡/⚪）。

写入

~/.claude/plugins/config/claude-for-legal/litigation-legal/matters/[slug]/chronology.md

（或根据标志使用格式变体）。

若存在旧版本：版本号递增，向用户展示差异摘要。
最终确认前提示："这是我构建的内容。请查看🔴标记的条目——有没有我误判的？"

Chronology

时间线构建

Disclosed-document use restrictions

披露文档使用限制

Before working with a set of litigation documents, ask: "Were any of these documents obtained through disclosure or discovery in legal proceedings?" If yes:

England & Wales (CPR 31.22): Documents obtained through disclosure are subject to the implied undertaking — you may only use them for the purpose of the proceedings in which they were disclosed, unless the court grants permission, the disclosing party consents, or the document has been read in open court. Using them for a different matter, a different claim, or a commercial purpose without permission is a contempt.
US: Protective orders and Rule 26(c) may impose similar restrictions. Check the order.
Other jurisdictions: Similar restrictions commonly apply. Check the local rule.

Confirm: "This use is within the proceedings in which the documents were disclosed, or I have permission / consent, or the documents are now public." If not confirmed, flag it: "⚠️ Disclosed documents may have use restrictions. Confirm this use is permitted before proceeding."

在处理诉讼文档集之前，需询问："这些文档是否是通过法律程序中的披露或取证获得的？"如果是：

英格兰及威尔士（CPR 31.22）：通过披露获得的文档受默示承诺约束——仅可用于披露所针对的诉讼程序，除非获得法院许可、披露方同意，或该文档已在公开法庭宣读。未经许可将其用于其他案件、其他索赔或商业目的属于藐视法庭行为。
美国：保护令和Rule 26(c)可能施加类似限制，请核查相关命令。
其他司法管辖区：通常也有类似限制，请核查当地规则。

需确认："本次使用符合文档披露所针对的诉讼程序，或我已获得许可/同意，或该文档现已公开。"若未确认，需标记："⚠️ 披露文档可能存在使用限制，请在继续前确认本次使用已获许可。"

Purpose

目的

Facts happen in order. The chronology is the spine every narrative hangs on — the statement of facts in a brief, reserve memos, settlement memos, depo prep, witness prep. Building a chron by hand is slow; AI is good at structured extraction. The catch: garbage-in, garbage-out. This skill pulls from the sources the configuration declares and from whatever the user uploads.

事实按顺序发生。时间线是所有叙事的核心——包括诉状中的事实陈述、储备备忘录、和解备忘录、证词准备、证人准备。手动构建时间线效率低下；AI擅长结构化提取。但需注意：输入垃圾，输出垃圾。本技能从配置指定的源和用户上传的内容中提取信息。

Modes

模式

This skill serves two practice settings. Pick a default from the user's

## Role

in the plugin's configuration CLAUDE.md; the user can override per-run with a flag.

--matter
mode (default for in-house litigation counsel). Matter-history-focused. Reads the matter's case theory and key facts from
```
matter.md
```
, pulls from declared document-storage sources (Google Drive, SharePoint, Gmail, iManage, CLM — whatever the
```
## Landscape
```
section of CLAUDE.md declares), and treats
```
history.md
```
as the running internal log (decisions, holds, reserve memos — intentionally not in the chronology). Output is matter-centric: what happened across the dispute, tagged for advocacy use.
--documents
mode (default for firm associate / paralegal). Production-document-focused. Reads the case theory from the configuration, then extracts from an eDiscovery export, a custodial file set, or a Bates-numbered production. Output is production-centric: what the documents show, with Bates citations, tagged per the case theory.

Both modes converge on the same output structure (timeline, 🔴/🟡/⚪ significance tags, gaps, SoF variant). The difference is the source profile and the significance frame.

## Role

solo

other

, default to

--matter

but mention both modes on the first run and let the user pick.

本技能适用于两种场景。根据插件配置CLAUDE.md中的

## Role

选择默认模式；用户可通过标志在每次运行时覆盖默认设置。

--matter
模式（企业诉讼法律顾问默认）：聚焦案件历史。从
```
matter.md
```
读取案件理论和关键事实，从指定的文档存储源（Google Drive、SharePoint、Gmail、iManage、CLM——CLAUDE.md的
```
## Landscape
```
部分指定的任何源）提取内容，并将
```
history.md
```
视为运行中的内部日志（决策、保留通知、储备备忘录——不会纳入时间线）。输出以案件为中心：记录纠纷中的所有事件，标记用于辩护用途。
--documents
模式（律所律师/律师助理默认）：聚焦产出文档。从配置中读取案件理论，然后从电子取证导出文件、保管人文件集或带Bates编号的产出文件中提取内容。输出以产出文件为中心：记录文档显示的内容，附带Bates引用，并根据案件理论标记。

两种模式的输出结构一致（时间线、🔴/🟡/⚪重要性标签、缺口、事实陈述变体）。差异在于源范围和重要性框架。

若

## Role

为

solo

或

other

，默认使用

--matter

模式，但首次运行时需提及两种模式并让用户选择。

Side framing (significance tags)

立场框架（重要性标签）

The same event is significant in different ways depending on whether the practitioner is proving a claim or disproving it. Read

## Side

in the practice profile (and the per-matter posture if the matter overrides the default):

Plaintiff (offensive framing) — 🔴 marks events that establish elements of the claim (liability, causation, damages, notice), close gaps the defense will try to open, or start statute-of-limitations clocks in the plaintiff's favor. 🟡 marks events that support the claim but are subject to impeachment. ⚪ is background context.
Defense (defensive framing) — 🔴 marks events that break elements of the claim (failure of causation, notice, reliance), open statute-of-limitations or jurisdictional defenses, or support affirmative defenses (release, waiver, assumption of risk, comparative fault). 🟡 marks events that undermine the plaintiff's narrative. ⚪ is background.
Both / varies — ask the user per-chronology which side's framing to apply for significance tags. The underlying timeline is side-neutral; only the significance read changes.

Note the applied framing at the top of the output:

Significance tags applied from [plaintiff / defense] perspective.

When producing a Statement of Facts variant, use the side default unless the user specifies otherwise.

同一事件的重要性因从业者是主张索赔还是反驳索赔而异。读取实践配置中的

## Side

（以及案件是否覆盖默认立场）：

原告（进攻性框架） —— 🔴标记那些确立索赔要素（责任、因果关系、损害、通知）、填补被告试图打开的缺口，或启动对原告有利的诉讼时效时钟的事件。🟡标记那些支持索赔但可能被弹劾的事件。⚪为背景信息。
被告（防御性框架） —— 🔴标记那些打破索赔要素（因果关系不成立、未通知、无信赖）、开启诉讼时效或管辖权抗辩，或支持积极抗辩（免责、弃权、风险承担、比较过错）的事件。🟡标记那些削弱原告叙事的事件。⚪为背景信息。
双方/不确定 —— 每次构建时间线时询问用户应采用哪一方的框架进行重要性标记。基础时间线中立于立场；仅重要性解读会变化。

在输出顶部标注所应用的框架：

重要性标签采用[原告/被告]视角。

生成事实陈述变体时，除非用户指定，否则使用默认立场。

Load context

加载上下文

Common:

Plugin configuration CLAUDE.md → case theory context (in-house:
```
## Landscape
```
for document sources; firm associate:
```
## Case theory
```
and
```
## Document review
```
for platform + custodians),
```
## Outputs
```
for the work-product header,
```
## Decision posture
```
for the privilege-flagging rule.
Prior
```
chronology.md
```
for this matter, if it exists.
Any files the user uploads or paths they provide in-session.

--matter

mode also reads:

~/.claude/plugins/config/claude-for-legal/litigation-legal/matters/[slug]/matter.md

→ case theory, key facts, pivot fact (for significance tagging), key dates.

Default matter folder pattern from CLAUDE.md → where docs for this slug live.

--documents

mode also reads:

eDiscovery platform metadata if a connector is available (Everlaw, Relativity, DISCO, Aurora) — by custodian + date range.
Bates-range manifest or production index if the user points at one.

Conflicts gate — unbypassable (
--matter
mode). Before building the chronology, check

~/.claude/plugins/config/claude-for-legal/litigation-legal/matters/_log.yaml

for the matter slug. If the matter is not in

_log.yaml

, refuse and route:

"I don't see [matter slug] in the matter log. Run
/litigation-legal:matter-intake
first so the conflicts check runs and the matter workspace is set up. I won't build a chronology on a matter that hasn't been intaken — the conflicts check is the gate."

Do not proceed on an unintaken matter. Intake is what runs conflicts and writes the

_log.yaml

row this skill reads from.

--documents

mode (running against an ad-hoc document set without a matter slug) is exempt from the gate, but its outputs should be treated as pre-matter research and not filed as if matter work product.

常见上下文：

插件配置CLAUDE.md → 案件理论上下文（企业场景：
```
## Landscape
```
用于文档源；律所律师场景：
```
## Case theory
```
和
```
## Document review
```
用于平台+保管人）、
```
## Outputs
```
用于工作产品页眉、
```
## Decision posture
```
用于特权标记规则。
若存在该案件的旧版
```
chronology.md
```
，需加载。
用户本次会话中上传的任何文件或提供的路径。

--matter

模式还需读取：

```
~/.claude/plugins/config/claude-for-legal/litigation-legal/matters/[slug]/matter.md
```
→ 案件理论、关键事实、核心事实（用于重要性标记）、关键日期。
CLAUDE.md中的默认案件文件夹规则 → 该slug对应的文档存储位置。

--documents

模式还需读取：

若有可用连接器，读取电子取证平台元数据（Everlaw、Relativity、DISCO、Aurora）——按保管人+日期范围提取。
若用户指定，读取Bates范围清单或产出索引。

冲突关卡——不可绕过（
--matter
模式）。构建时间线前，检查

~/.claude/plugins/config/claude-for-legal/litigation-legal/matters/_log.yaml

中的案件slug。若案件不在

_log.yaml

中，需拒绝并引导：

"我在案件日志中未找到[案件slug]。请先运行
/litigation-legal:matter-intake
，以便进行冲突检查并设置案件工作区。我不会为未录入的案件构建时间线——冲突检查是必经关卡。"

不得为未录入的案件继续操作。录入流程会运行冲突检查并写入本技能需读取的

_log.yaml

条目。

--documents

模式（针对无案件slug的临时文档集运行）可豁免此关卡，但其输出应视为案件前研究，不得作为案件工作产品归档。

Workflow

工作流程

Step 0: Privilege gate (runs first, every time)

步骤0：特权审查关卡（每次运行首先执行）

Chronology work pulls from documents. Documents are often privileged (attorney-client, work product, common interest, joint defense) — in-house matter files often are by default; eDiscovery productions, especially rolling productions or common-interest productions, often contain privileged or unreviewed material. Extracting content from a privileged document into a chronology that later gets shared can risk waiver, depending on who receives it and under what doctrine (common-interest, joint-defense, Kovel, and work-product protections may apply). Waiver analysis is fact-specific — get counsel sign-off before distributing.

The skill will not extract until the user picks a privilege posture:

Before I extract: how have the sources been privilege-screened?
A. All sources cleared — you've already screened these. I extract without privilege flags. Output is discovery-ready posture; still marked work product.
B. Mixed or not yet screened — I extract and tag every entry with a
priv
flag:
ok
(sourced from clearly non-privileged material),
flag
(sourced from potentially privileged material — A/C, WP, common interest), or
review
(source unclear). Flagged entries are visually marked in the output, and the Statement-of-Facts variant filters them out by default.
C. Abort — screen first — pause the skill. Screen the sources. Return and re-run.

Record the choice in the chronology header as

privilege_posture: A-cleared | B-mixed | C-aborted

. If B or C, record the rationale briefly.

Why a gate and not just a warning: a warning gets read once and forgotten. A gate forces the posture decision into the record, which means every chronology file carries its own provenance — anyone reading it later knows whether entries were derived from privilege-screened material.

时间线构建需从文档提取内容。文档通常享有特权（律师-客户保密、工作产品、共同利益、联合辩护）——企业案件文件通常默认享有特权；电子取证产出文件，尤其是滚动产出或共同利益产出文件，通常包含享有特权或未审查的材料。从享有特权的文档中提取内容到时间线并后续共享，可能导致特权放弃，具体取决于接收方和适用原则（共同利益、联合辩护、Kovel和工作产品保护可能适用）。特权放弃分析基于具体事实——分发前需获得律师签字确认。

本技能需用户选择特权立场后方可提取：

在提取前，请告知这些源的特权审查情况：
A. 所有源已审查通过 —— 您已完成审查。我将提取内容且不标记特权。输出符合取证就绪标准；仍标记为工作产品。
B. 混合或未审查 —— 我将提取内容并为每个条目标记
priv
标志：
ok
（来源为明确非特权材料）、
flag
（来源为潜在特权材料——律师-客户保密、工作产品、共同利益）或
review
（来源不明确）。标记的条目会在输出中可视化标注，且事实陈述变体默认过滤掉这些条目。
C. 终止——先审查 —— 暂停本技能。请先审查源。返回后重新运行。

在时间线页眉中记录选择：

privilege_posture: A-cleared | B-mixed | C-aborted

。若选择B或C，简要记录理由。

为何设置关卡而非仅警告：警告只会被读取一次并遗忘。关卡会强制将立场决策记录在案，意味着每个时间线文件都带有自身来源——后续读取者可明确知晓条目是否来自已审查特权的材料。

Step 1: Identify document sources

步骤1：识别文档源

--matter
mode:

User-provided paths — anything dropped in this session (file paths, drive links, email exports).
Default matter folder — from CLAUDE.md's document-storage pattern, expanded for this slug (e.g.,
```
G:/Legal/Matters/acme-v-us-2026
```
).
Declared sources — the
```
Document storage
```
table in CLAUDE.md, filtered to ones this matter might touch (e.g., Gmail archive for sender-side communications, SharePoint legal folder).
Ask — if sources look thin, prompt: "I can build from what I have, but the chronology will be incomplete. Anything else to point me at? Key emails, contracts, internal memos, production letters?"

--documents
mode:

Production export / Bates set — the user points at the production directory or a manifest; the skill reads by Bates range + date.
eDiscovery connector — if an MCP connector is available (Everlaw, Relativity, DISCO, Aurora), pull by custodian + date range.
Custodial files — if the user provides raw custodial mailboxes or drive exports, read those too.
Ask — if coverage looks thin for a key custodian or date range, prompt.

--matter
模式：

用户提供的路径 —— 本次会话中上传的任何内容（文件路径、云端链接、邮件导出文件）。
默认案件文件夹 —— 从CLAUDE.md的文档存储规则扩展到该slug（例如：
```
G:/Legal/Matters/acme-v-us-2026
```
）。
指定源 —— CLAUDE.md中的「文档存储」表，筛选该案件可能涉及的源（例如：发件方通信的Gmail存档、SharePoint法律文件夹）。
询问 —— 若源内容不足，提示："我可以基于现有内容构建，但时间线会不完整。是否有其他内容可以提供？例如关键邮件、合同、内部备忘录、产出函件？"

--documents
模式：

产出导出文件/Bates集 —— 用户指定产出目录或清单；技能按Bates范围+日期读取。
电子取证连接器 —— 若有可用MCP连接器（Everlaw、Relativity、DISCO、Aurora），按保管人+日期范围提取。
保管人文件 —— 若用户提供原始保管人邮箱或云端导出文件，也需读取。
询问 —— 若关键保管人或日期范围的覆盖不足，提示用户补充。

Step 2: Pull + read

步骤2：提取+读取

For each source with readable files:

PDFs, emails (.eml), .docx, .txt — read directly.
Email archives (Gmail, Outlook) — if an MCP connector is authenticated, query by date range + counterparty / key terms; otherwise the user exports relevant threads to a folder.
eDiscovery platforms (Everlaw, Relativity, DISCO, Aurora) — if connector is available, pull by custodian + date range; otherwise the user provides an export.

If the skill can't access a declared source, name it explicitly in the output's Gaps section rather than silently proceeding.

No silent supplement. If source coverage for an era of the matter is thin — fewer documents than expected for a claimed time window, a custodian whose mailbox isn't accessible, a production that hasn't landed — report what was found and stop. Do NOT fill gaps from web search, public record search, or model knowledge about the matter without asking. Say: "Sources returned [N] events for [period / custodian]. Coverage appears thin. Options: (1) point me at additional sources (Bates, folder, mailbox), (2) try a different MCP connector if configured, (3) search the web for public-record events in this window — results will be tagged

[web search — verify]

and should be checked against a primary source before relying, or (4) stop here and note the gap. Which would you like?" A lawyer decides whether to accept lower-confidence sources; the skill does not decide for them.

Source attribution. Tag every chronology entry with where the event came from: the file path, Bates number, MCP connector, or declared document-storage source for events extracted from retrieved documents (already captured in the Sources column). For any event or date that cannot be traced to a retrieved document — e.g., a fact recalled from model training data, a public-record event found via web search — tag it inline:

[web search — verify]

[model knowledge — verify]

, or

[user provided]

where the user stated the fact in-session. Entries tagged

verify

carry higher fabrication risk than document-sourced entries and should be checked first. Never strip or collapse the tags — they are counsel's fastest signal about which entries to verify before pulling them into a brief or SoF.

Tagging reaches every section that states a legal conclusion, deadline, or computed date — not just timeline entries. The timeline is sourced from documents. The Gaps section, the Key events section, the Theory tie lines, and any statement of limitations, tolling event, filing deadline, discovery cutoff, or privilege determination are legal analysis the skill writes from model knowledge unless sourced. Every such statement carries a provenance tag:

[computed from: <rule cited with tag>]

[model knowledge — verify]

[user provided]

, or a research-connector tag if retrieved in this session. A statute-of-limitations window with no tag defaults to

[model knowledge — verify]

. A "key event" line that characterizes a fact's legal significance is analysis and needs the tag. The rule is simple: if it's an assertion about the law, not an assertion about what a document says, it must carry the same provenance tag the timeline entries do. When no research connector is reachable and the skill is computing deadlines or citing rules, record it in the Sources: line of the reviewer note (see plugin CLAUDE.md

## Outputs

) — do not emit a standalone banner.

对于每个包含可读取文件的源：

PDF、邮件(.eml)、.docx、.txt —— 直接读取。
邮件存档（Gmail、Outlook） —— 若已认证MCP连接器，按日期范围+交易方/关键词查询；否则用户需将相关线程导出到文件夹。
电子取证平台（Everlaw、Relativity、DISCO、Aurora） —— 若有可用连接器，按保管人+日期范围提取；否则用户需提供导出文件。

若技能无法访问指定源，需在输出的「缺口」部分明确标注，不得静默继续。

不得静默补充。若案件某一时期的源覆盖不足——某一时间段的文档数量少于预期、保管人邮箱无法访问、产出文件未到——需报告已发现的内容并停止。未经询问，不得通过网络搜索、公开记录搜索或模型对案件的认知填补缺口。需提示："源在[时间段/保管人]返回[N]个事件。覆盖不足。选项：(1) 提供额外源（Bates编号、文件夹、邮箱），(2) 尝试配置的其他MCP连接器，(3) 搜索该时间段的公开记录事件——结果会标记

[web search — verify]

，依赖前需核对原始源，或(4) 停止并标注缺口。请选择？"律师有权决定是否接受低置信度源；技能不得自行决定。

源归因。为每个时间线条目标记事件来源：从检索文档提取的事件需标注文件路径、Bates编号、MCP连接器或指定的文档存储源（已在「源」列捕获）。对于任何无法追溯到检索文档的事件或日期——例如模型训练数据中的事实、网络搜索发现的公开记录事件——需内联标记：

[web search — verify]

、

[model knowledge — verify]

或

[user provided]

（若用户在会话中陈述该事实）。标记

verify

的条目比文档源条目具有更高的伪造风险，应优先核查。不得删除或合并这些标记——这是律师快速判断哪些条目需在纳入诉状或事实陈述前核查的信号。

标记需覆盖所有陈述法律结论、截止日期或计算日期的部分——不仅限于时间线条目。时间线源自文档。「缺口」部分、「关键事件」部分、「理论关联」以及任何关于诉讼时效、中断事件、提交截止日期、取证截止日期或特权认定的陈述，除非有来源，否则均为技能从模型知识生成的法律分析。每个此类陈述都需带有来源标记：

[computed from: <引用规则的标签>]

、

[model knowledge — verify]

、

[user provided]

，或本次会话中检索的研究连接器标签。无标记的诉讼时效窗口默认标记为

[model knowledge — verify]

。描述事实法律重要性的「关键事件」条目属于分析，需标记。规则很简单：如果是关于法律的断言，而非关于文档内容的断言，必须带有与时间线条目相同的来源标记。当无法访问研究连接器且技能需计算截止日期或引用规则时，需在审阅者备注的「源：」行记录（见插件CLAUDE.md的

## Outputs

）——不得单独显示横幅。

Step 3: Extract events

步骤3：提取事件

For each document, identify dated events:

Email:

[date] [sender] told [recipient] [subject/content]

Meeting:
```
[date] [attendees] met about [topic]
```
(per calendar entry or notes)
Decision:
```
[date] [decision-maker] decided [what]
```
(per memorializing doc)

Filing / pleading:

[date] [party] filed [motion/complaint/response]

External event:
```
[date] [thing happened]
```
(contract signed, product launched, regulator acted, event crossed a threshold)

One event per document usually. Occasionally zero (undated or no event established). Sometimes multiple (meeting summary covering several decisions).

Privilege flag per entry (only when privilege_posture == B-mixed). Three-state rule — never silently decide a subjective privilege test isn't met:

```
priv: ok
```
— source is confidently non-privileged (filings, regulatory correspondence, public docs, counterparty communications without our counsel). Used only when there's no plausible privilege theory.
```
priv: flag
```
— source is confidently or likely privileged (communications with counsel, work-product memos, privileged drafts, joint-defense material). Default for anything uncertain — if the dominant-purpose call is close, or litigation contemplation is borderline, or the content is mixed, it goes here, not in
```
ok
```
.
```
priv: review
```
— source unclear on its face, but the skill could not make the call at all (no sender/recipient metadata, unreadable, etc.).

When

priv: flag

priv: review

, add

[SME VERIFY: privilege status]

inline so the counsel sees it during review. Under-flagging waives privilege (one-way door); over-flagging is corrected by counsel in review (two-way door). Prefer the recoverable error.

对于每个文档，识别带日期的事件：

邮件：

[日期] [发件人]告知[收件人][主题/内容]

会议：

[日期] [参会者]就[主题]召开会议

（根据日历条目或纪要）

决策：
```
[日期] [决策者]决定[事项]
```
（根据记录文档）

提交/诉状：

[日期] [当事人]提交[动议/起诉状/答辩状]

外部事件：
```
[日期] [事件内容]
```
（合同签署、产品发布、监管行动、事件达到阈值）

通常每个文档对应一个事件。偶尔为零（无日期或未确立事件）。有时多个（会议纪要涵盖多项决策）。

每个条目标记特权（仅当privilege_posture == B-mixed时）。三态规则——不得静默判定主观特权测试不满足：

```
priv: ok
```
—— 来源明确非特权（提交文件、监管通信、公开文档、无我方律师参与的交易方通信）。仅当无合理特权理论时使用。
```
priv: flag
```
—— 来源明确或可能享有特权（与律师的通信、工作产品备忘录、特权草稿、联合辩护材料）。不确定时默认使用——若主要目的判定模糊、诉讼考量边界不清，或内容混合，标记为此类，而非
```
ok
```
。
```
priv: review
```
—— 来源表面不明确，技能无法做出判定（无发件人/收件人元数据、无法读取等）。

当标记为

priv: flag

或

priv: review

时，需内联添加

[SME VERIFY: privilege status]

，以便律师在审阅时看到。标记不足会导致特权放弃（单向门）；标记过度可通过律师审阅纠正（双向门）。优先选择可挽回的错误。

Step 4: De-dupe

步骤4：去重

The same event surfaces in multiple documents: a meeting is on three calendars and produces a summary email — that's one event with four sources, not four events. Merge. The merged entry cites all sources.

同一事件可能出现在多个文档中：某会议出现在三个日历中并生成了纪要邮件——这是一个事件对应四个源，而非四个事件。需合并。合并后的条目需引用所有源。

Step 5: Tag significance — per case theory

步骤5：标记重要性——根据案件理论

Read the pivot fact and key facts from

matter.md

(

--matter

mode) or from the configuration's

## Case theory

section (

--documents

mode). Tag each event:

🔴 Key — event is part of the pivot fact or a key fact for/against us
🟡 Relevant — context, pattern evidence, supports a secondary argument
⚪ Background — useful for completeness, not going in the brief

Discipline: a chronology of 300 entries with 300 🔴 tags has no tags. Reserve 🔴 for events that would genuinely move a factfinder. If in doubt, 🟡.

Borderline tagging: when an entry sits between 🔴 and 🟡 (or 🟡 and ⚪), tag at the lower significance and add

[SME VERIFY — borderline significance call]

inline. Counsel's judgment will override the skill's call. A chronology that confidently over-tags is less useful than one that surfaces its uncertainty.

从

matter.md

（

--matter

模式）或配置的

## Case theory

部分（

--documents

模式）读取核心事实和关键事实。为每个事件标记：

🔴 关键 —— 事件属于核心事实或对我方有利/不利的关键事实
🟡 相关 —— 背景信息、模式证据、支持次要论点
⚪ 背景 —— 有助于完整性，但不会纳入诉状

原则：300个条目全部标记🔴的时间线等于没有标记。仅将🔴保留给真正能影响事实认定者的事件。若不确定，标记为🟡。

边界标记：当条目介于🔴和🟡（或🟡和⚪）之间时，标记为较低重要性并内联添加

[SME VERIFY — borderline significance call]

。律师的判断会覆盖技能的判定。明确过度标记的时间线不如暴露不确定性的时间线有用。

Step 6: Write

步骤6：写入

Default output is the working chronology. Variants on request.

默认输出为工作用时间线。可根据请求生成变体。

Output formats

输出格式

Working chronology (default)

工作用时间线（默认）

Location:

~/.claude/plugins/config/claude-for-legal/litigation-legal/matters/[slug]/chronology.md

. Complete, tagged, annotated. The reference doc counsel works from.

markdown

[WORK-PRODUCT HEADER — per plugin config ## Outputs — differs by role; see `## Who's using this`]

> **Privilege inheritance.** This chronology is derived from matter documents that may be attorney-client-privileged, work-product-protected, common-interest / joint-defense material, or a mix. It inherits the sources' protection status. Distributing it beyond the privilege circle — to business stakeholders outside the engagement, to opposing counsel, to a regulator — can waive protection over both the chronology and the underlying sources. Store with privileged matter material, mark consistently with house privilege conventions, and make distribution decisions deliberately. The privilege-posture choice captured below is the provenance stamp for any later distribution call.

位置：

~/.claude/plugins/config/claude-for-legal/litigation-legal/matters/[slug]/chronology.md

。完整、带标记、带注释。是律师工作的参考文档。

markdown

[工作产品页眉 — 根据插件配置## Outputs — 因角色而异；见`## Who's using this`]

> **特权继承**。本时间线源自可能享有律师-客户保密特权、工作产品保护、共同利益/联合辩护材料的案件文档，或混合类型文档。它继承了源的保护状态。将其分发到特权圈之外——例如参与业务的外部利益相关者、对方律师、监管机构——可能导致时间线和底层源的保护被放弃。请与特权案件材料一起存储，按照律所特权惯例统一标记，并谨慎做出分发决策。下方记录的特权立场选择是后续分发决策的来源标记。

Chronology — [Matter Name]

时间线 — [案件名称]

Significance tags (🔴/🟡/⚪) and privilege flags (🔒) are first-pass reads requiring
[SME VERIFY]
before use in any external work product (briefs, SoF, board memo, outside counsel deliverable).

Matter: [slug] Mode: matter | documents Built: [YYYY-MM-DD] Sources: [N] documents across [source types] Entries: [N] ([N] 🔴 / [N] 🟡 / [N] ⚪) Pivot fact: [one sentence] Privilege posture: A-cleared | B-mixed | C-aborted Flagged entries: [N] 🔒 (only present when posture == B-mixed)

重要性标签（🔴/🟡/⚪）和特权标志（🔒）为初步判定，在用于任何外部工作产品（诉状、事实陈述、董事会备忘录、外部律师交付件）前需
[SME VERIFY]
。

案件： [slug] 模式： matter | documents 构建时间： [YYYY-MM-DD] 源： [N]份文档，涵盖[源类型] 条目： [N]条（[N]条🔴 / [N]条🟡 / [N]条⚪） 核心事实： [一句话] 特权立场： A-cleared | B-mixed | C-aborted 标记条目： [N]条🔒 (仅当立场为B-mixed时显示)

Timeline

时间线

Date	Event	Tag	🔒	Sources
[YYYY-MM-DD]	[what happened, one sentence]	🔴/🟡/⚪	[blank / 🔒-flag / 🔒-review]	[file paths or Bates]

日期	事件	标签	🔒	源
[YYYY-MM-DD]	[事件内容，一句话]	🔴/🟡/⚪	[空白 / 🔒-flag / 🔒-review]	[文件路径或Bates编号]

Key events (🔴 only)

关键事件（仅🔴）

[Pulled out, each with a line on why it matters to the theory.]

[提取出来，每个事件附带说明其对理论的重要性。]

[date] — [event title]

[日期] — [事件标题]

What: [one line]
Theory tie: [why this matters]
Sources: [list]

内容：[一句话]
理论关联：[重要性原因]
源：[列表]

Gaps

缺口

Date ranges with no events: [ranges — where are documents for this period?]

Expected but missing: [events we'd expect to see documented but don't — e.g., "contract amendments between 2024-06 and 2025-03 — not produced"]

Unreadable sources: [sources declared in CLAUDE.md but not accessible this run — e.g., "Everlaw production — no MCP connector; export needed"]

无事件的日期范围： [范围 — 该时间段的文档在哪里？]

预期但缺失的事件： [我们预期会有文档记录但未找到的事件 — 例如："2024-06至2025-03期间的合同修正案 — 未产出"]

无法读取的源： [CLAUDE.md中指定但本次运行无法访问的源 — 例如："Everlaw产出文件 — 无MCP连接器；需导出"]

Marker discipline

标记规则

[VERIFY: factual assertion — date, attendees, content]

— not yet confirmed against the underlying doc

[UNCERTAIN: legal characterization — e.g., whether an event establishes a regulatory trigger]

[CITE NEEDED: Bates / exhibit / depo page:line]

[SME VERIFY: privilege status | borderline significance call]

— counsel judgment needed

[VERIFY: factual assertion — date, attendees, content]

— 尚未根据底层文档确认

[UNCERTAIN: legal characterization — e.g., whether an event establishes a regulatory trigger]

[CITE NEEDED: Bates / exhibit / depo page:line]

[SME VERIFY: privilege status | borderline significance call]

— 需要律师判断

Version

版本

v[N] built on [date] from [source summary]
v[N-1] built on [date] (prior, superseded)

undefined

v[N] 构建于[日期]，源自[源摘要]
v[N-1] 构建于[日期]（旧版，已取代）

undefined

Statement-of-facts chronology (on request)

事实陈述时间线（按需生成）

Filter to 🔴 and relevant 🟡 only. Present as prose in chronological narrative order — the skeleton for a brief's fact section. Each paragraph is one event or tightly linked cluster, with record citations.

Privilege filter default: when

privilege_posture == B-mixed

, 🔒-flagged and 🔒-review entries are excluded by default. The SoF variant is intended for eventual external use (briefs, disclosures, negotiating counterparty) — 🔒 entries don't belong there until counsel confirms privilege status. If the user wants 🔒 entries included anyway, require explicit

--include-flagged

acknowledgment; capture the acknowledgment in the output header as permanent record.

仅筛选🔴和相关🟡条目。按时间顺序以叙事散文形式呈现——作为诉状事实部分的框架。每个段落对应一个事件或紧密关联的事件群，附带记录引用。

特权过滤默认规则：当

privilege_posture == B-mixed

时，🔒-flag和🔒-review条目默认排除。事实陈述变体用于最终外部用途（诉状、披露、与对方谈判）——🔒条目在律师确认特权状态前不应纳入。若用户希望纳入🔒条目，需明确确认

--include-flagged

；并在输出页眉中记录该确认作为永久记录。

Witness-specific chronology (on request)

特定证人时间线（按需生成）

Filter to events where a named witness is sender, recipient, attendee, or subject. Feeds witness prep and helps reconstruct what a witness knew when.

筛选指定证人作为发件人、收件人、参会者或主题的事件。用于证人准备，帮助重建证人在特定时间的认知。

Incremental builds

增量构建

chronology.md

exists:

Read prior version
Build new chronology from current sources
Diff: new events (since last build), modified entries (new sources added to existing events), removed entries (rare; note why)
Preserve the prior version number; write new version with
```
v[N+1]
```
Output summary of what changed

若

chronology.md

已存在：

读取旧版本
根据当前源构建新时间线
差异：新增事件（自上次构建以来）、修改条目（现有事件添加新源）、删除条目（罕见；需说明原因）
保留旧版本号；写入新版本
```
v[N+1]
```
输出变更摘要

Integration with matter.md / history.md

与matter.md / history.md的集成

Intentionally separate (in-house

--matter

mode).

history.md

is counsel's running log — decisions, updates, procedural milestones, internal strategy notes.

chronology.md

is the advocacy-facing timeline of facts. They overlap but don't merge:

A hold was issued → goes in history.md (internal action). Usually not in chronology (not a fact of the dispute).
The counterparty sent a breach notice on March 14 → goes in chronology.md (🟡 — establishes their knowledge). Also in history.md if the intake referenced it.
Our reserve recommendation memo was drafted → history.md only.

When counsel wants history events in the chronology, they can paste them. The default is they stay separate.

故意分离（企业

--matter

模式）。

history.md

是律师的运行日志——决策、更新、程序里程碑、内部战略笔记。

chronology.md

是面向辩护的事实时间线。两者重叠但不合并：

发出保留通知 → 纳入history.md（内部行动）。通常不纳入时间线（不属于纠纷事实）。
对方于3月14日发送违约通知 → 纳入chronology.md（🟡 — 确立其知晓情况）。若录入时提及，也会纳入history.md。
我方储备建议备忘录起草完成 → 仅纳入history.md。

当律师希望将history事件纳入时间线时，可粘贴内容。默认情况下两者分离。

What this skill does not do

本技能不执行的操作

Resolve contradictions. When two documents say different things about when an event happened, both entries go in with a flag. Resolution is counsel's call; may require witness interview or further discovery.
Invent events not in the sources. If it's not in the documents (and not in matter.md or the configuration as a captured fact), it's not in the chronology — but "Gaps" might call it out as missing.
Guarantee completeness. A chronology is only as good as the sources. If the eDiscovery production is ongoing and only 20% has landed, the chronology reflects that. Name the limitation.
Decide privilege status for the user. The Step 0 gate forces the posture choice; the per-entry
```
priv
```
flag captures first-pass classification. Actual privilege determinations are counsel's call per
```
[SME VERIFY]
```
flags.

解决矛盾。当两个文档对事件发生时间有不同表述时，两个条目都需纳入并标记。解决矛盾是律师的职责；可能需要证人访谈或进一步取证。
编造源中不存在的事件。若文档中不存在（且未在matter.md或配置中作为已捕获事实记录），则不纳入时间线——但「缺口」部分可能会标注为缺失。
保证完整性。时间线的质量取决于源。若电子取证产出正在进行中，仅到账20%，时间线会反映这一点。需标注局限性。
为用户判定特权状态。步骤0的关卡强制用户选择立场；每个条目的
```
priv
```
标志记录初步分类。实际特权认定需律师根据
```
[SME VERIFY]
```
标志做出。