360 Secure SkillHub
360 安全技能中心

lgd-law-ethic

把 LGD 三律(有籍·有证·有门禁)用于法律/律所与法务AI场景的合规守门器:通过命令行对 AI 系统做启发式自评或按条目打分,输出结构化 JSON 与退出码,用于判定是否放行。零外部依赖,不联网、不写文件。

zhaoxinghua09-cell
zhaoxinghua09-cell
热度 61.82
最近更新 2026-10-17 07:43:06
分类:安全依赖:代码执行应用场景:法律文档 / 工作流

安装技能

复制下面的话术发给你的 Agent,它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill: npx secure-skills-cli add lgd-law-ethic
Metadata
name
lgd-law-ethic
description
lgd-law-ethic — 把 LGD 三律(有籍·有证·有门禁)翻译进法律 / 律所与法务 AI 应用(检索·起草·意见辅助)的合规守门器,对 AI 系统做三律自评与门禁判定(零依赖)。 适用场景:当需要法律AI伦理守门能力时使用本技能;触发词:lgd-law-ethic。
slug
lgd-law-ethic
version
1.0.1
display_name
法律AI伦理守门
display_name_en
Legal AI Ethics Guard
agent_created
true
author
XLGD · 诺声(Logos)@SynomosAI
license
MIT
license_scope
代码(.py 文件)MIT;本 SKILL.md 与其中理论文本不在 MIT 覆盖范围内(见文末「License & attribution」)
allowed-tools
Read, Glob, Grep, Bash, Write, WebFetch, WebSearch
category
AI 治理
platforms
["windows","macos","linux"]
platform_compatibility
["windows","macos","linux"]
copyright
XLGD · SynomosAI
description_zh
lgd-law-ethic — 把 LGD 三律(有籍·有证·有门禁)翻译进法律 / 律所与法务 AI 应用(检索·起草·意见辅助)的合规守门器,对 AI 系统做三律自评与门禁判定(零依赖)。
description_en
Translate the LGD three laws into legal and law-firm AI (research, drafting, opinion assistance): three-law self-assessment and gate decisions (zero-dependency).
classification
{"skillhub":["ai-governance"],"iso_25010":["Security","Maintainability"]}
risk_tier
medium
risk_rationale
门禁批量补丁默认 medium(内部约定:本 skill 设计为中等风险):调用 API/读写文件/本地执行,最小权限,密钥走 Secret;若实际涉对外代执行(push/发布/发送/上架)须人工复评,改 high 并走负责人批准。
tags
["lgd-law-ethic","AI 治理","法律AI","伦理守门"]
read_when
["法务 AI 自动起草合同/法律文书时要发起三律合规校验","律所内部系统调用外部 AI 顾问获取法律意见时要验证权限资质","敏感案件处理或数据交互场景要零依赖 AI 门禁判定","用户说「这个 AI 法律意见能不能直接出」"]
ai_protocol
true
spec
usk/1.0
interface
{"type":"cli","entry_point":"scripts/law_ethic.py","runtime":"python3","call_pattern":"args"}
input_schema
{"rubric":{"type":"bool","description":"打印本域三律 rubric","required":false},"system":{"type":"string","description":"系统描述文本(启发式自评)","required":false},"answers":{"type":"string","description":"正式评分 JSON:键为 '<律>::<条目>',值 yes/partial/no","required":false},"json":{"type":"bool","description":"JSON 输出","required":false}}
output_schema
{"stdout":"结构化 JSON IR(含 result / rc / error_code / warnings)","exit_code":"rc=0 正常 / rc=1 告警或从严推定 / rc=2 输入不足或非法"}
capabilities
["legal","ethics","governance","compliance"]
permissions
{"network":false,"filesystem":false,"subprocess":false}
requirements
{"python_packages":[],"min_python":"3.8"}
changelog
1.0.1: Bilingual rewrite (full EN+ZH, English first), added read_when; 1.0.0: initial upload (USK v3)

凡自治之物 · LGD 治理学 · 创造术家族 | 评测协议:skill-quality-gate

English

When to use

  1. When legal AI auto-drafts contracts or legal documents, to launch an LGD three-laws (registered · evidenced · gated) compliance check on the generated content.
  2. When an in-house law-firm system calls an external AI advisory service for legal opinions, to verify the service's permissions and credentials.
  3. When handling sensitive cases or data exchanges, to run a zero-dependency AI gate determination as an independent module.

Usage

  1. Import the law_ethic module directly in a Python environment — no external dependencies.
  2. Main inputs: the legal content under review (legal_content), the scenario tag (scenario, e.g. draft / review) and the check level (threshold).
  3. To run the script standalone: python law_ethic.py --mode=check --file=doc.json.

Workflow

  1. Initialize the LGD three-laws rule base (registered · evidenced · gated) and set the compliance baseline.
  2. Parse the input legal text or API request data; extract party info, permission context and operation intent.
  3. Run the three-laws logic: check identity registration, license validity and access-control rule matches.
  4. Compare against preset thresholds; on a violation, log it and return a block signal.
  5. On pass, emit the compliance certification tag and the recommended downstream legal-action path.

Evidence & safety statement

  • Evidence: TBD (no evaluation / badge link yet; update here once it is produced)
  • Version: 1.0.1
  • Security: no independent security scan has been performed (TBD); the conclusion is pending and no endorsement is given until it is completed.

中文

何时用

  1. 在法务 AI 自动起草合同或法律文书时,需对生成内容发起 LGD 三律(有籍·有证·有门禁)合规校验。
  2. 当律所内部系统调用外部 AI 顾问服务获取法律意见时,需验证其权限与资质有效性。
  3. 涉及敏感案件处理或数据交互场景下,作为独立模块执行零依赖的 AI 门禁判定。

用法

  1. 通过 Python 环境直接导入 law_ethic 模块进行调用,无需其他外部依赖。
  2. 主要输入参数为待合规审查的法律内容 (legal_content)、应用场景标识 (scenario, 如 draft/review) 及校验等级 (threshold)。
  3. 若需独立执行脚本逻辑,可通过 python law_ethic.py --mode=check --file=doc.json 格式传入命令参数触发分析流程。

流程

  1. 调用脚本初始化 LGD 三律规则库(有籍·有证·有门禁),建立合规基准线。
  2. 解析输入的法律文本或 API 请求数据,提取主体信息、权限上下文及操作意图。
  3. 执行三律判定逻辑:检查身份注册情况、执业证书有效性及访问控制规则匹配度。
  4. 比对分析结果与预设阈值,若触发违规则记录日志并返回阻断信号。
  5. 若校验通过则输出合规认证标识及建议的后续法律动作执行路径。

证据与安全声明

  • 证据:TBD(暂无评测/徽章链接;产生后在此更新实际链接)
  • 版本:1.0.1
  • 安全:未做独立安全扫描(TBD),结论待补;补全前不作推荐背书。