Skip to content
Back to skills

Clause Risk Review Assistant

ASecurity

用于合同条款风险审查,基于关键词与规则识别高风险表述(单方决定、责任失衡、模糊期限、无限责任等)并输出修订建议。适用于法务风控预审、合同谈判前风险筛查场景。

  • 16 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added June 8, 2026
ai-agentspythonbashgit

Security analysis

A100/100

Scanned June 8, 2026

npx -y skills add ThomasMoreAI/legal-skills-open --skill clause-risk-review-assistant --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Clause Risk Review Assistant?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Clause Risk Review Assistant
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/thomasmoreai-clause-risk-review-assistant/badge)](https://www.skillsdirectory.com/skills/thomasmoreai-clause-risk-review-assistant)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: clause-risk-review-assistant
title: 条款风险审查助手(T107)
description: 用于合同条款风险审查,基于关键词与规则识别高风险表述(单方决定、责任失衡、模糊期限、无限责任等)并输出修订建议。适用于法务风控预审、合同谈判前风险筛查场景。
author: aifinlab
author_url: https://github.com/aifinlab/FinClaw/tree/main/skills/clause-risk-review-assistant
license: Apache-2.0
version: 0.1.0
execution_mode: open
jurisdiction: general
practice: contracts
language: zh
---

# 条款风险审查助手(T107)

## 概述

本技能用于识别合同条款中的常见风险表达,输出风险等级与修订建议,辅助法务快速聚焦重点条款。

## 输入要求

支持 JSON 数组或 JSONL,每条记录建议包含:

- `contract_name`
- `clause_no`
- `clause_title`
- `clause_text`

## 工作流

1. 汇总待审查条款
2. 运行脚本进行风险扫描
3. 对高风险条款补充法律判断
4. 输出风险清单与修订建议
5. 跟踪修订闭环

## 脚本调用指引

使用 `scripts/clause_risk_review.py` 生成报告:

```bash
python scripts/clause_risk_review.py \
  --input clauses.json \
  --output clause_risk_report.md
```

可选参数:
- `--rules`:自定义风险规则 JSON
- `--top`:展示高风险条款数量(默认 20)

## 输出结构

1. 风险等级统计
2. 高风险条款清单
3. 风险命中依据与建议修改方向
4. 待律师复核事项
5. 方法与限制

## 质量要求

- 每个风险命中必须给出触发依据
- 修订建议需具体可执行
- 不输出法律结论,仅输出审查建议

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…