shield_lock
JinDaGe
← Back to skills
extension
Category: Security & ComplianceNo API key required

defense lawyer

AI-powered criminal defense assistant for Chinese legal system - provides case analysis, defense strategy formulation, evidence assessment, and legal documen...

personAuthor: zhenstaffhubclawhub
open_in_newRepositorydownloadDownload package

Defense Lawyer - AI辩护律师助手

English | 中文

<a name="english"></a>

English

AI-powered criminal defense assistant for Chinese legal system

Version: 0.1.0 | License: MIT

What is Defense Lawyer?

Defense Lawyer is an AI-powered assistant designed specifically for criminal defense in the Chinese legal system. It provides comprehensive support for lawyers and legal professionals in case analysis, strategy formulation, and legal document generation.

Core Features

  • Case Analysis - Comprehensive case evaluation with conviction and sentencing risk assessment
  • Defense Strategy - Formulate innocence defense, lesser crime defense, or sentencing defense strategies
  • Evidence Assessment - Analyze authenticity, legality, and relevance of evidence (three-property test)
  • Legal Research - Retrieve laws, judicial interpretations, guiding cases, and precedents
  • Document Generation - Auto-generate defense statements, legal opinions, and appeal petitions
  • Sentencing Analysis - Identify mitigating factors and assess probation possibilities

Quick Start

Installation

# Install from source
pip install openclaw-defense-lawyer

# Or clone repository
git clone https://github.com/ZhenRobotics/openclaw-defense-lawyer.git
cd openclaw-defense-lawyer
pip install -e .

Basic Usage

import asyncio
from datetime import datetime
from defense_lawyer import DefenseLawyer, CriminalCase, Client, Evidence

async def main():
    # Initialize defense lawyer assistant
    lawyer = DefenseLawyer()

    # Define case information
    case = CriminalCase(
        case_id="2024-CASE-001",
        case_name="Zhang Theft Case",
        case_type="Theft",
        charge="Theft Crime",
        incident_date=datetime(2024, 1, 15),
        incident_location="Beijing Chaoyang District Mall",
        case_description="Defendant stole a mobile phone worth 5,000 RMB...",
        proceeding_stage="First instance",
        statutory_penalty="Up to three years imprisonment, criminal detention, or public surveillance",
    )

    # Define client information
    client = Client(
        name="Zhang",
        gender="Male",
        age=28,
        detention_status="Arrested",
        confession_status="Plea bargaining",
        remorse_level="Deep remorse",
        compensation_made=True,
        victim_forgiveness=True,
    )

    # Define evidence
    evidences = [
        Evidence(
            evidence_id="E001",
            evidence_type="Audio-visual material",
            evidence_name="Mall surveillance video",
            relevance="Direct evidence",
            authenticity="Authentic",
            legality="Legal",
            probative_value="Strong",
        ),
    ]

    # 1. Case analysis
    analysis = await lawyer.analyze_case(case, client, evidences)
    print(f"Conviction risk: {analysis.conviction_risk}")
    print(f"Sentencing risk: {analysis.sentencing_risk}")

    # 2. Formulate defense strategy
    strategy = await lawyer.formulate_defense_strategy(case, client, evidences)
    print(f"Strategy type: {strategy.strategy_type}")

    # 3. Sentencing analysis
    sentencing = await lawyer.analyze_sentencing(case, client)
    print(f"Probation possibility: {sentencing.suspended_sentence_possibility}")

    # 4. Generate defense statement
    doc = await lawyer.generate_defense_statement(case, client, strategy)
    with open("defense_statement.md", "w", encoding="utf-8") as f:
        f.write(doc.markdown_content)

asyncio.run(main())

Use Cases

Scenario 1: Initial Case Assessment

assessment = await lawyer.quick_assessment(case, client)
print(f"Conviction risk: {assessment['conviction_risk']}")
print(f"Recommended strategy: {assessment['recommended_strategy']}")

Scenario 2: Trial Preparation

package = await lawyer.complete_defense_package(case, client, evidences)
# Includes: case analysis, strategy, sentencing analysis, defense statement, legal research

Scenario 3: Evidence Cross-Examination

for evidence in prosecution_evidences:
    assessment = await lawyer.assess_evidence(evidence)
    print(f"Score: {assessment.overall_score}/100")
    print(f"Cross-examination strategies: {assessment.challenge_strategies}")

Testing

# Run complete test suite
python3 test_complete.py

# Run example
python3 example.py

Test Coverage: 7/7 modules passing (100%)

Python Advantages

| Feature | Python | JavaScript | |---------|--------|------------| | PDF Processing | ⭐⭐⭐⭐⭐ | ⭐⭐ | | Document Generation | ⭐⭐⭐⭐⭐ | ⭐⭐⭐ | | Data Analysis | ⭐⭐⭐⭐⭐ | ⭐⭐ | | Legal Databases | ⭐⭐⭐⭐⭐ | ⭐⭐⭐ |

Disclaimer

What Defense Lawyer Can Do:

  • Provide professional case analysis and recommendations
  • Assist in formulating defense strategies
  • Generate legal document drafts
  • Provide legal research and case references

What Defense Lawyer Cannot Do:

  • Replace professional legal judgment
  • Guarantee specific litigation outcomes
  • Provide formal legal opinions
  • Replace attorney courtroom defense

Important: This system serves as a legal assistance tool only. All outputs require review by licensed attorneys before use.

Roadmap

v0.2.0 (2026 Q2)

  • Real legal database API integration
  • Support for 50+ common crimes
  • Civil case support
  • LaTeX format document output

v0.3.0 (2026 Q3)

  • Multi-jurisdiction legal differences
  • Case similarity analysis
  • Visualization of sentencing charts
  • Team collaboration features

License

MIT License - See LICENSE

Links

  • GitHub: https://github.com/ZhenRobotics/openclaw-defense-lawyer
  • Issues: https://github.com/ZhenRobotics/openclaw-defense-lawyer/issues
  • Author: Justin Wang (code@zhenrobot.com)

<a name="chinese"></a>

中文

AI驱动的刑事辩护专业助手 - 为中国法律体系量身定制

版本: 0.1.0 | 许可证: MIT

Defense Lawyer 是什么？

Defense Lawyer 是一个专为刑事辩护设计的AI助手系统，针对中国法律体系，为律师和法律工作者提供全面的案件分析、辩护策略制定和法律文书生成支持。

核心功能

  • 案件分析 - 全面的案件事实总结、法律适用分析、定罪/量刑风险评估
  • 辩护策略 - 制定无罪辩护、罪轻辩护、量刑辩护等多种策略
  • 证据评估 - 分析证据的真实性、合法性、关联性（证据三性）
  • 法律研究 - 检索相关法律法规、司法解释、指导性案例和判例
  • 文书生成 - 自动生成辩护词、法律意见书、上诉状等法律文书
  • 量刑分析 - 识别从轻减轻情节，分析缓刑可能性，提出量刑建议

快速开始

安装

# 从 PyPI 安装
pip install openclaw-defense-lawyer

# 或者克隆仓库
git clone https://github.com/ZhenRobotics/openclaw-defense-lawyer.git
cd openclaw-defense-lawyer
pip install -e .

基础用法

import asyncio
from datetime import datetime
from defense_lawyer import DefenseLawyer, CriminalCase, Client, Evidence

async def main():
    # 初始化辩护律师助手
    lawyer = DefenseLawyer()

    # 定义案件信息
    case = CriminalCase(
        case_id="2024-刑初-001",
        case_name="张某盗窃案",
        case_type="盗窃罪",
        charge="盗窃罪",
        incident_date=datetime(2024, 1, 15),
        incident_location="北京市朝阳区某商场",
        case_description="被告人张某在商场盗窃手机一部，价值人民币5000元...",
        proceeding_stage="一审阶段",
        statutory_penalty="三年以下有期徒刑、拘役或者管制",
    )

    # 定义当事人信息
    client = Client(
        name="张某",
        gender="男",
        age=28,
        detention_status="逮捕",
        confession_status="认罪认罚",
        remorse_level="深刻悔罪",
        compensation_made=True,
        victim_forgiveness=True,
    )

    # 定义证据
    evidences = [
        Evidence(
            evidence_id="E001",
            evidence_type="视听资料",
            evidence_name="商场监控视频",
            relevance="直接证据",
            authenticity="真实",
            legality="合法",
            probative_value="强",
        ),
    ]

    # 1. 案件分析
    analysis = await lawyer.analyze_case(case, client, evidences)
    print(f"定罪风险: {analysis.conviction_risk}")
    print(f"量刑风险: {analysis.sentencing_risk}")

    # 2. 制定辩护策略
    strategy = await lawyer.formulate_defense_strategy(case, client, evidences)
    print(f"策略类型: {strategy.strategy_type}")

    # 3. 量刑分析
    sentencing = await lawyer.analyze_sentencing(case, client)
    print(f"缓刑可能性: {sentencing.suspended_sentence_possibility}")

    # 4. 生成辩护词
    doc = await lawyer.generate_defense_statement(case, client, strategy)
    with open("辩护词.md", "w", encoding="utf-8") as f:
        f.write(doc.markdown_content)

asyncio.run(main())

使用场景

场景1：初次会见后的案情评估

assessment = await lawyer.quick_assessment(case, client)
print(f"定罪风险: {assessment['conviction_risk']}")
print(f"建议策略: {assessment['recommended_strategy']}")

场景2：准备庭审材料

package = await lawyer.complete_defense_package(case, client, evidences)
# 包含：案件分析、辩护策略、量刑分析、辩护词、法律研究

场景3：证据质证准备

for evidence in prosecution_evidences:
    assessment = await lawyer.assess_evidence(evidence)
    print(f"评分: {assessment.overall_score}/100")
    print(f"质证策略: {assessment.challenge_strategies}")

测试

# 运行完整测试套件
python3 test_complete.py

# 运行示例
python3 example.py

测试覆盖: 7/7 模块全部通过 (100%)

Python 优势

| 功能 | Python | JavaScript | |------|--------|------------| | PDF 处理 | ⭐⭐⭐⭐⭐ | ⭐⭐ | | 法律文书生成 | ⭐⭐⭐⭐⭐ | ⭐⭐⭐ | | 数据分析 | ⭐⭐⭐⭐⭐ | ⭐⭐ | | 法律数据库 | ⭐⭐⭐⭐⭐ | ⭐⭐⭐ |

免责声明

Defense Lawyer 能做什么：

  • 提供专业的案件分析和辅助建议
  • 协助制定辩护策略和方案
  • 生成法律文书草稿
  • 提供法律研究和判例参考

Defense Lawyer 不能做什么：

  • 替代专业律师的法律判断
  • 保证任何具体的诉讼结果
  • 提供正式的法律意见
  • 替代律师的庭审辩护

重要提示：本系统仅作为法律工作辅助工具，所有输出内容需要专业律师审核后使用。

发展路线图

v0.2.0（2026年第二季度）

  • 集成真实的法律法规数据库API
  • 增加更多罪名支持（50+常见罪名）
  • 支持民事案件分析
  • LaTeX 格式法律文书输出

v0.3.0（2026年第三季度）

  • 多地区法律差异支持
  • 判例相似度智能分析
  • 可视化量刑分析图表
  • 团队协作功能

许可证

MIT License - 详见 LICENSE

链接

  • GitHub: https://github.com/ZhenRobotics/openclaw-defense-lawyer
  • 问题反馈: https://github.com/ZhenRobotics/openclaw-defense-lawyer/issues
  • 作者: Justin Wang (code@zhenrobot.com)

为法律工作者赋能 ⚖️

让AI成为每位辩护律师的得力助手