All skills
cat-xierluo avatar

/legal-skill-alignment

@9be736d

写法律 Skill 前的前置对齐步骤——通过苏格拉底式提问把零散的法律经验/素材/直觉厘清为结构化 Brief(legal-skill-brief/v1),交给 skill-creator 编译。适用于律师想把文书经验、办案 SOP、咨询记录做成 Skill 的场景。不适用于非法律类 Skill 创建,也不适用于用户已提供完整可执行 Brief 且明确要求直接编译的情况。

Use this Skill: https://skilld.dev/gh/cat-xierluo/legal-skills/legal-skill-alignment

This session only. Nothing lands on disk.

evalscase-bundle-260728v1-realignedcase-01-contract-reviewmaterials.md

≈534 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Case 01 Materials: 供应商服务合同审查

v1 重对齐案例材料清单。本案例对应 v0 评测 case-bundle-260705/case-01-contract-review/ 的重对齐版(参见 v0 EVALUATION.md 作为基线)。

素材类型(v1 七类分类)

类型 数量 来源 quality 用途
sample(成品文书) 2 历史审查意见书(仓储配送服务协议 + SaaS 服务协议) silver 提取结构骨架 / 风格特征
sop(流程规则) 1 律所内部供应商审查清单(8 步骤) gold 提取决策点 / 阈值
qa 0 无 unrated(无 Q&A 素材) —
rules(规则包) 1 SOP 中包含的阈值表 gold 阈值与判断标准
revisions 0 无 unrated(无对话修订记录) —
tools-data 0 无 unrated —
authorities 0 无(从样本中反向提取法条,见 Brief) unrated(待与原文核对) 法源验证

v0 → v1 差异(与 case-bundle-260705/case-01 对照)

  • 素材分类:v0 仅 sample × 2 + sop × 1;v1 显式标为 sample / sop / rules / authorities 四类。
  • quality 字段:v0 未填;v1 全填(silver / gold / unrated)。
  • authorities 单独列出:v1 起法源作为独立素材类型管理,便于回溯验证状态。
  • 三角色拆分:v0 笼统写"终端用户角色:企业法务部 + 采购部门";v1 拆为 operator = buyer-counsel、represented_party = buyer、output_audience = buyer-internal-team。
  • 法源三列:v0 法源条目无 jurisdiction / effective_date / verification_status;v1 全部补充。
  • 阈值 source 列:v0 阈值无来源;v1 全部带 source(SOP v3 / sample-002 反推)。
  • 安全段:v0 无;v1 必填,含敏感信息脱敏状态与外传策略。
  • 待确认分类:v0 用 - [ ] 列表;v1 按 blocker / warning 分类。
  • 下游契约:v0 移交语 "请调用 legal-skill-creator 编译";v1 改为中立 "可交给通用 skill-creator 或 legal-skill-creator"。

Source: SKILL.md on GitHub

1 warning10d3 checks · Risk SAFE
  • Gen Agent Trust Hub10d

    The analysis of the legal-skill-alignment skill reveals a robust security posture focused on the structured alignment of legal knowledge without performing high-risk operations. The skill primarily serves as an instructional framework for generating structured markdown briefs, incorporating significant internal quality controls and de-sensitization protocols. No evidence of malicious patterns such as remote code execution, persistence, or credential exfiltration was detected. Potential risks related to indirect prompt injection are addressed through specific instructions for data de-identification and the use of 'unverified' status for unconfirmed legal sources. The included evaluation documentation demonstrates a high level of testing rigor and integrity checking.

  • Socket10d

    No alerts

  • Snyk10d

    Risk: MEDIUM · 1 issue

Signed by skilld at 9be736d. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated 2 weeks ago
homepage
https://github.com/cat-xierluo/legal-skills
author
杨卫薪律师(微信ywxlaw)
version
1.0.7

README badge

README badge for cat-xierluo/legal-skills/legal-skill-alignment