mcpField
M
memo

memo

github.com/anthropics文档处理9.8k632 次安装14 周前更新
代码仓库

关于 memo

面向法律诊所学生的案件分析技能,根据案情搭建 IRAC 框架并标出研究缺口。它保留分析和结论供学生完成,并依照导师设定的教学模式调整引导方式。

主要功能

  1. 01按法律问题分别搭建 IRAC
  2. 02标注待研究规则与法律问题
  3. 03列出事实应用提示,留空分析和结论
  4. 04整理案件强弱项及待确认问题
  5. 05依据导师设置调整教学方式

适用场景

  1. 01根据接案摘要和笔记准备案件分析备忘录
  2. 02识别尚需法律研究或补充事实的部分
  3. 03为与教授讨论准备开放问题和策略议题

安装

复制下面的命令,在终端中运行。

$npx skills add https://github.com/anthropics/claude-for-legal --skill memo

安装后在 Claude Code 中输入/memo确认已加载。

详细介绍

/memo

  1. 加载 ~/.claude/plugins/config/claude-for-legal/legal-clinic/CLAUDE.md → 执业领域、司法管辖区。
  2. 使用以下工作流。阅读接案摘要/案件笔记。
  3. 将争议点表述为问题。为每个问题搭建 IRAC 框架——规则部分标记为需研究,分析部分提供学生分析提示,结论部分留空。
  4. 总结优势/劣势/待解问题。总结研究缺口。
  5. 输出时显著标注“分析由你完成”。
/legal-clinic:memo

备忘录:案件内部分析

目的

案件分析备忘录是学生展现思考的地方。此技能提供 IRAC 框架并指出研究缺口——由学生完成分析。

分析由学生完成。 此技能负责搭建结构,不负责下结论。

加载背景信息

~/.claude/plugins/config/claude-for-legal/legal-clinic/CLAUDE.md → 执业领域、司法管辖区、指导风格。 接案摘要和案件笔记用于了解事实。

教学方式检查

阅读该执业领域的督导指南:~/.claude/plugins/config/claude-for-legal/legal-clinic/guides/<practice-area>.md。检查 pedagogy_posture 设置:

  • guide(默认): 提供 IRAC 结构和研究缺口列表。要求学生自行通过研究起草每条规则陈述,而不是直接提供框架。对学生写的内容提供反馈。只有在学生尝试过一次后,才提出为某一部分补充框架规则。
  • assist: 提供备忘录框架,并填写能够填写的内容。标出需要学生审阅的项目。学生通过编辑和审阅来学习。(注意:此备忘录技能始终会按设计将 [STUDENT ANALYSIS] 和 [STUDENT CONCLUSION] 部分留空——assist 表示技能提供 IRAC 框架和框架规则陈述;不会代写分析或结论。)
  • teach: 不提供框架或框架内容。要求学生提出争议点、陈述其研究所得的规则,并进行分析。然后提供反馈。学生遇到困难时,提出引导性问题。只有在学生尝试两次后,才提供规则陈述范例或分析段落范例,并且仅针对学生遇到困难的部分。记录学生答对和答错的内容,以便督导了解其进展。

如果没有指南,则使用 guide。如果指南存在但未设置教学方式,则使用 guide。

无论采用哪种教学方式,输出中始终包含:“教学模式:[assist/guide/teach] — 由督导的指南设定。这意味着我[说明学生完成了什么,以及此技能完成了什么]。”

工作流

第 1 步:提出争议点

根据接案摘要和案件笔记,确定本案涉及哪些法律问题。

将每个问题表述为疑问句。不要只写“适居性”,而应写“客户能否以暖气损坏为由提出适居性抗辩来反对驱逐?如果可以,该抗辩能否抵扣所欠租金?”

如果存在多个问题,则每个问题都应有各自的 IRAC 部分。

Step 2:搭建 IRAC 框架

针对每个问题:

问题: 以问句形式陈述(来自 Step 1)。

规则: 这是一个研究缺口,而非结论。说明学生需要查明什么:

[RESEARCH NEEDED: [State] habitability doctrine — warranty of habitability elements, what conditions qualify, remedies available including rent offset. Start with: [State] landlord-tenant statute, then case law on heater/heat specifically. See /research-start for a roadmap.]

如果该技能对一般规则框架有较高把握(例如“多数州承认默示的适居性保证”),可以将其作为框架起点陈述——但必须明确标注为未经核实:

框架(未经核实——请确认是否适用于 [State]): 多数法域承认默示的适居性保证,要求房东维持适宜人类居住的条件。违反该保证可能产生扣留租金、维修后抵扣租金或减租的救济。 [VERIFY: [State]'s specific elements and remedies]

适用: 学生在此进行分析。搭建结构,不要代为填写:

`[STUDENT ANALYSIS: Apply the rule to the facts. Key facts to address:

  • Heater broken since November — how long is "unreasonable"?
  • Client notified landlord [when? how? documented?]
  • Landlord's response or lack thereof
  • [State]-specific: does client need to have given written notice? deposited rent in escrow? other procedural prerequisites?]`

列出重要事实。由学生自行进行适用分析。

结论: 明确留空:

[STUDENT CONCLUSION: Based on your research and analysis above, what's the likely outcome? How strong is this defense? What are the weaknesses?]

Step 3:找出优势、弱点和待解决的问题

在 IRAC 模块之后单独设置一个部分:

优势(从事实中显而易见——学生应验证):

  • [看似有利的事实及其原因]

弱点(从事实中显而易见——学生应评估其严重程度):

  • [看似不利的事实及其原因]
  • [UNCERTAIN: whether [X] is actually a weakness — depends on [State] rule on [Y]]

待解决的问题(备忘录在缺少更多信息时无法回答的问题):

  • 事实方面:[尚未从客户处了解的信息]
  • 法律方面:[需要研究的内容]
  • 策略方面:[由学生/教授作出判断的问题]

输出

═══════════════════════════════════════════════════════════════════════
  AI-ASSISTED SCAFFOLD — THE ANALYSIS IS YOURS TO WRITE
  Every [RESEARCH NEEDED] and [STUDENT ANALYSIS] block is a prompt, not
  a placeholder to delete. The thinking happens when you fill them in.
═══════════════════════════════════════════════════════════════════════

# Case Analysis Memo: [Client] — [Matter]

**Date:** [date] | **By:** [student] | **For:** [Professor]

---

## Bottom line

[Take the case / Decline because X / Need more info on Y — next step is Z]

---

## Issues Presented

1. [Issue as question]
2. [Issue as question]

---

## Issue 1: [Issue]

### Rule

[Framework starting point with VERIFY flags, and RESEARCH NEEDED blocks]

### Application

[STUDENT ANALYSIS scaffold with the facts that matter]

### Conclusion

[STUDENT CONCLUSION — blank]

---

[repeat for each issue]

---

## Strengths

[list with caveats]

## Weaknesses

[list with UNCERTAIN flags where applicable]

## Open Questions

**Factual:** [list]
**Legal:** [list — these feed /research-start]
**Strategic:** [list — these are for discussion with Professor]

---

## Research gaps summary

[Every RESEARCH NEEDED block pulled out into one list, so the student can
work through them systematically — and can run /research-start on each]

═══════════════════════════════════════════════════════════════════════

## What this memo is NOT

This is a scaffold, not an analysis. The [STUDENT ANALYSIS] blocks are where
the educational value lives — filling them in is the work. A memo where those
blocks are still empty is a memo that hasn't been written yet.

---

**Cite verification — required before use.** Any framework rules, cases, or statutes suggested above were generated by an AI model and have not been verified. Before relying on any citation — or including it in client work — run it through Westlaw, CourtListener, or your clinic's research platform for accuracy and current good-law status. Flag unverified citations to your supervisor.

**Source attribution.** Tag every suggested citation in the scaffold with where it came from: `[Westlaw]`, `[CourtListener]`, or the MCP tool name for citations retrieved from a legal research connector; `[web search — verify]` for web-search citations; `[model knowledge — verify]` for citations recalled from training data; `[user provided]` for citations the supervising attorney or case file supplied. Citations tagged `verify` carry higher fabrication risk than tool-retrieved citations and should be checked first. Never strip or collapse the tags — they are the supervisor's fastest signal about which citations to verify.

**No silent supplement.** If a query to a configured research tool returns few or no results for a rule the memo needs, say so and stop. Do NOT fill the gap from web search or model knowledge without asking. Say: "The search returned [N] results from [tool]. Coverage appears thin for [rule / issue]. Options: (1) broaden the search query, (2) try a different research tool, (3) search the web — results will be tagged `[web search — verify]` and should be checked against a primary source before relying, or (4) leave `[RULE TO VERIFY]` and stop. Which would you like?" The supervising attorney decides whether to accept lower-confidence sources.

此技能不会做什么

  • 代写分析。 它会搭建 IRAC 框架并标出缺口,由学生推理并进行适用分析。
  • 提供已核实的规则。 每条规则陈述在学生完成研究前都会明确标注为未经核实。
  • 得出结论。 IRAC 中的 C 刻意留空。
  • 取代与教授的交流。 “待解决的问题/策略”部分是与教授交流的议程,而非替代品。

以“下一步决策树”收尾

按照 CLAUDE.md ## Outputs 中的要求,以“下一步决策树”收尾。根据此技能刚刚生成的内容定制选项——五个默认分支(起草 X、升级处理、获取更多事实、观察等待、其他事项)只是起点,并非固定选项。决策树就是输出;由律师作出选择。

评论

支持 Markdown