← Back to Skills Marketplace
210
Downloads
0
Stars
1
Active Installs
4
Versions
Install in OpenClaw
/install judgment-analyzer
Description
分析判决书文档，提取关键信息并生成结构化分析报告。当用户提到「分析判决书」「解读判决书」「判决书总结」「生成判决书报告」「批量分析判决书」时使用。支持 PDF 和 Word (.docx/.doc) 格式，可处理单个文件或整个文件夹。
README (SKILL.md)
判决书分析技能
分析本地判决书文档，提取案件关键信息，生成结构化的摘要报告和综合分析报告。
工作流程
步骤 1：接收输入
用户指定判决书路径，支持：
- 单文件：
/path/to/judgment.pdf - 文件夹：
/path/to/cases/（批量处理所有 PDF/Word 文件） - 支持格式：PDF、Word (.docx/.doc)
步骤 2：文本提取（Python）
运行 scripts/analyzer.py 提取原始文本：
python3 ~/.claude/skills/judgment-analyzer/scripts/analyzer.py \x3C输入路径>
脚本会：
- 读取每个判决书文件（PDF 或 Word）
- 提取纯文本内容
- 将文本保存到
输入路径/摘要/文件夹，文件名为原文件名.txt
步骤 3：生成摘要（AI）
- 读取
摘要/文件夹中的所有 txt 文件 - 按照输出格式模板，对每个案件生成结构化摘要
- 将摘要保存为
案件名_摘要.md - 批量处理时，额外生成
综合分析报告.md
输出格式要求
在提取信息和生成报告时，必须严格遵守模板文件中的格式和要求设置。
具体格式以及各项指引，请读取并遵循此文件：
~/.claude/skills/judgment-analyzer/references/output-template.md
- 单案件摘要：遵照模板文件中【摘要报告模式】的 6 个核心模块提取并输出。
- 批量综合报告：遵照模板文件中【综合分析报告模式】的结构和表格要求输出。
触发方式
- 「分析判决书 /path/to/case.pdf」
- 「批量分析 /Users/xxx/Desktop/cases/」
- 「生成判决书综合报告」
- 「对比分析这几份判决书」
注意事项
- 脚本路径：
~/.claude/skills/judgment-analyzer/scripts/analyzer.py - 首次使用需安装依赖：
pip install -r ~/.claude/skills/judgment-analyzer/scripts/requirements.txt - 批量处理时，自动跳过非判决书文件（.pdf, .docx, .doc）
- 摘要文件夹创建在输入文件夹同级目录
- 优先提取文本，如文本质量差可标记"扫描件，需OCR"
Usage Guidance
This skill appears coherent and focused: it extracts text locally and asks the AI to generate reports from those texts. Before installing/running: (1) review the included scripts (analyzer.py) yourself — it only reads/writes local files and uses pdfplumber/python-docx; (2) run dependency installation in an isolated environment (virtualenv) to limit impact of PyPI packages; (3) be aware the skill will read any files you point it at and will write .txt outputs into a '摘要' folder next to your inputs, so avoid pointing it at sensitive system directories; (4) if you need OCR for scanned PDFs, the skill notes optional OCR deps but does not automatically install them—review and install only what you trust. Overall no credentials or network endpoints were found and the behavior aligns with the stated purpose.
Capability Analysis
Type: OpenClaw Skill
Name: judgment-analyzer
Version: 1.2.1
The 'judgment-analyzer' skill is designed to extract and analyze text from legal judgment documents (PDF/Word). The Python script (analyzer.py) uses standard libraries (pdfplumber, python-docx) to perform local text extraction and saves the results to a local directory, while the SKILL.md and output-template.md provide structured instructions for the AI to summarize the findings. No evidence of data exfiltration, malicious execution, or harmful prompt injection was found.
Capability Assessment
Purpose & Capability
Name/description (判决书分析) matches the provided artifacts: a Python extractor (pdfplumber, python-docx), a template, and SKILL.md describing text extraction + AI-driven analysis. Nothing requested (no env vars, no binaries) is unrelated to the stated purpose.
Instruction Scope
SKILL.md instructs running the included analyzer.py to read local PDF/Word files, write .txt outputs, then have the AI read those txt files and produce reports per the local template. The instructions only reference local paths under the skill folder and the user-specified input path; they do not ask for other system files, credentials, or remote endpoints.
Install Mechanism
This is instruction-only with an included script; there is no automated install spec. The README/SKILL.md ask the user to run `pip install -r requirements.txt` which pulls libraries from PyPI (pdfplumber, python-docx). Installing Python packages from PyPI is normal but has the usual supply-chain considerations (review packages, use virtualenv).
Credentials
The skill requires no environment variables, no credentials, and no config paths. It only reads user-specified local files and writes extracted text to a sibling '摘要' folder—this is proportionate to its purpose.
Persistence & Privilege
Flags are default (always: false, user-invocable: true). The skill does not request persistent system-wide privileges or modify other skills' configurations. Autonomous model invocation is allowed by platform default but is not combined with broad credentials or other red flags.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install judgment-analyzer - After installation, invoke the skill by name or use
/judgment-analyzer - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.2.1
v1.2.1: 更新技能显示名称，增加中文标识。
v1.2.0
v1.2.0: 合并争议焦点与法院认定，增强逻辑穿透深度。
v1.1.0
Refactored output templates and extraction logic
v1.0.0
Initial release of judgment-analyzer skill.
- Analyze local court judgment documents (PDF, Word) to extract key case information and generate structured summary reports.
- Supports single file and batch (folder) processing, including output of per-case summaries and a consolidated analysis report.
- Automated extraction of essential fields: case number, court, cause, date, parties, procedure, verdict, and key reasoning.
- Provides analysis guidelines based on case types (e.g., finance, tort, labor, IP, family).
- Offers ready-to-use output templates ensuring content completeness and consistency.
- Example usage and step-by-step workflow included for user clarity.
Metadata
Frequently Asked Questions
What is Judgment Analyzer 判决书分析?
分析判决书文档，提取关键信息并生成结构化分析报告。当用户提到「分析判决书」「解读判决书」「判决书总结」「生成判决书报告」「批量分析判决书」时使用。支持 PDF 和 Word (.docx/.doc) 格式，可处理单个文件或整个文件夹。 It is an AI Agent Skill for Claude Code / OpenClaw, with 210 downloads so far.
How do I install Judgment Analyzer 判决书分析?
Run "/install judgment-analyzer" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Judgment Analyzer 判决书分析 free?
Yes, Judgment Analyzer 判决书分析 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Judgment Analyzer 判决书分析 support?
Judgment Analyzer 判决书分析 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Judgment Analyzer 判决书分析?
It is built and maintained by szzzcode (@szzzcode); the current version is v1.2.1.
More Skills