SuperAPI AI Directory

legal-job-search

Office & productivity

View repository

Updated 10-16 08:10

About this skill

Legal AI Job Search Assistant

Overview

This skill helps legal professionals (in-house counsel, lawyers) use AI to improve efficiency and preparation quality during the job search process.

Core Capabilities:

  • In-depth company/law firm research (supports Deep Research fallback to Web Search)
  • Legal risk analysis (supports tool-level MCP fallback to Web Search)
  • Web resume generation (fully configurable style system)
  • Targeted resume adjustment
  • Interview memo generation

Work Path:

环境检测 → 信息收集 → 材料生成 → 输出确认

Workflow

Phase 0: Environment Detection and Path Planning

Step 0.1: Detect Available MCP Tools

Adoptcapability matchingstrategy, without limiting to specific tool names:

for 每个功能需求:
    │
    ├─ 功能1:企业被执行信息
    │   └─ 匹配关键词:executed, execution, enforcement, 被执行, 执行
    │
    ├─ 功能2:企业失信信息
    │   └─ 匹配关键词:dishonest, breach, 失信, 违约
    │
    ├─ 功能3:限制高消费
    │   └─ 匹配关键词:consumption, restriction, 限制高消费, 限消
    │
    ├─ 功能4:终本案件
    │   └─ 匹配关键词:final, closed, 终本
    │
    └─ 功能5:裁判文书
        └─ 匹配关键词:judgment, court, ruling, 裁判文书, 判决

匹配依据:
├─ 工具名称包含关键词
├─ 工具描述包含功能说明
└─ 工具参数包含 entname 或 company_name

Record the list of available tools.

For detailed instructions, see:references/mcp-tools/mcp-catalog.md

Step 0.2: Ask About Deep Research Tools

Ask the user whether they have the following tools:

  • Metaso AI Search (Deep Research mode)
  • Perplexity Pro
  • Other Deep Research tools

Step 0.3: Determine Job Type

岗位类型判断:
│
├─ 法务岗位 → 使用「公司调研框架」
├─ 律所岗位 → 使用「律所调研框架」
└─ 其他法律岗位 → 询问用户使用哪个框架

Phase 1: Information Collection

1.1 Job Information Extraction

Extract from the content provided by the user:

  • Job Title
  • Job Responsibilities
  • Job Requirements
  • Preferred Qualifications

If the user only provides the company/law firm name, guide the user to provide JD information.

1.2 Company/Law Firm Research

In-house Counsel Position - Company Research

优先路径:Deep Research 工具
│
├─ 秘塔AI搜索:开启「深度研究」模式
│   └─ 查询:"{公司名} 公司介绍 商业模式 业绩 行业地位"
│
├─ Perplexity Pro:
│   └─ 同样查询
│
└─ 降级路径:Web Search(多轮)
    ├─ 搜索1:{公司名} + "公司介绍" + "商业模式"
    ├─ 搜索2:{公司名} + "行业地位" + "竞争对手"
    ├─ 搜索3:{公司名} + "业绩" + "年报"
    └─ 整合生成简版报告

Law Firm Position - Law Firm Research

公开信息:
├─ 律所官网团队介绍
├─ 律师专业领域公示
├─ 律所业绩展示
└─ 行业排名/奖项

私域信息(用户自行补充):
├─ 使用 references/frameworks/firm-research-framework.md
└─ 提供问题清单引导用户收集

1.3 Legal Risk Analysis (Capability-level Fallback)

Execute independently for each capability requirement:

功能需求列表:
├─ 企业被执行信息
├─ 企业失信信息
├─ 限制高消费
├─ 终本案件
└─ 裁判文书

for 每个功能需求:
    │
    ├─ 尝试查找功能匹配的 MCP 工具
    │   ├─ 检查工具名称、描述、参数
    │   ├─ 找到匹配工具 → 尝试调用
    │   │   ├─ 调用成功 → 使用 MCP 数据
    │   │   └─ 调用失败 → 继续尝试其他工具
    │   │
    │   └─ 没有找到匹配工具 → 执行降级
    │
    └─ 降级:Web Search(多轮)
        ├─ 使用预设查询模板
        ├─ 多轮搜索获取信息
        └─ 整合结果到报告

For detailed instructions, see:references/frameworks/legal-risk-framework.md │ └─ Fallback: Web Search └─ Use the corresponding query templates in references/mcp-tools/fallback-queries.md └─ Integrate search results into the report


**输出结构**:
```markdown
## 法律风险分析报告

### 执行信息
- 查询时间:{时间戳}
- 数据来源:{MCP工具名称 / Web Search}

### 风险概览
| 风险类型 | 风险等级 | 核心问题 |
|---------|---------|---------|
|         |         |         |

### 详细信息
#### 被执行信息
...

#### 失信信息
...

(其他类别)

Phase 2: Basic Material Preparation

2.1 User Material Collection Guidance

First-time User Guidance:

If the user is using this for the first time or their materials are incomplete, guide the user to refer to the material preparation checklist:

请参考「法律人求职材料准备清单」提供你的信息。

清单位置:references/prompts/preparation-checklist.md

你可以:
1. 复制清单中的模板填写
2. 或直接按你自己的方式提供信息

Material Checklist:

必需信息:
├─ 基本信息(姓名、职位、年龄、学历、联系方式)
├─ 教育背景(学校、学历、专业、时间)
├─ 工作经历(公司、职位、时间、主要成就)
├─ 核心技能(专业技能、语言能力)
└─ 资格证书(如有)

推荐信息:
├─ 目标公司/律所名称
├─ 目标岗位 JD
└─ 自我评价

可选信息:
├─ 照片
├─ 项目经验
├─ 发表文章/论文
└─ 其他补充信息

2.2 Resume Information Processing

If the user already has a web resume:

  • Directly load the HTML file
  • Extract resume content for subsequent adjustments

If the user provides a Word/PDF resume:

  • Use docx skill or mineru-ocr skill to extract content
  • Organize into structured information
    • Educational Background
    • Work Experience
    • Core Skills
    • Qualifications
    • Self-Evaluation

2.2 Resume Style Selection

样式选择:
│
├─ 使用预设样式:
│   ├─ business-elite - 商务精英风(金色+深蓝)
│   ├─ modern-tech - 现代科技风(蓝色+灰白)
│   └─ minimal-clean - 极简清新风(绿色+米白)
│
├─ 或自定义样式:
│   └─ 使用 references/styles/custom-guide.md
│       └─ 用户指定:配色方案、字体、布局
│
└─ 或保持已有简历样式

Phase 3: Targeted Material Generation

3.1 Adjustment Strategy Development

Analyze the following content:

  • Keywords in the JD
  • Business characteristics of the company/law firm
  • Match points between the user's experience and the position

Output adjustment strategy:

# 简历调整策略

## 关键词匹配
| JD关键词 | 简历对应位置 |
|---------|-------------|
|         |             |

## 经验重构
- 前置:{需要前置的经验}
- 扩充:{需要扩充的经验}
- 合并:{可以合并的经验}

## 技能排序
按与岗位相关性重新排序:
1. {最相关技能}
2. {次相关技能}
...

3.2 Targeted Resume Generation

Use references/prompts/html-resume-template.md Prompt template in

  • Apply adjustment strategy
  • Match target style
  • Generate new HTML resume

3.3 Interview Memo Generation

Use the memo section in references/prompts/materials-prompt-template.md :

  • Company/Law Firm Core Information Overview
  • Business/Legal Interface Map
  • Risk and Challenge Analysis
  • Position Value Proposition
  • Interview Talking Points (Insights + Questions to Ask)

Output format: Polished Word document (.docx)

Use the docx skill to generate a Word document in a fixed format; for style specifications, see: references/prompts/memo-word-spec.md

Document Structure:

  • Cover page (company/law firm name, position name, preparer, date)
  • Table of contents page
  • Main body (five chapters)
  • Appendix (data source description, usage recommendations)

Style Features:

  • A4 page, professional color scheme (gold + dark blue)
  • Unified font specifications (Microsoft YaHei headings + SimSun body text)
  • Polished table styles (dark header background + zebra striping)
  • Headers and footers (chapter navigation + page numbers)

Phase 4: Output and Confirmation

4.1 File Output

输出文件:
├── 简历_{姓名}_{目标公司/律所}.html
├── 面试备忘录_{目标公司/律所}.docx  (精美 Word 文档)
└── 法律风险分析报告_{目标公司/律所}.md

4.2 (Optional) Interactive Resume Integration

If the user asks how to add a chat window:

  • Guide to extensions/interactive-resume/dify-integration.md

Dependencies

Required

  • Claude Code (supports HTML generation, Web Search)
  • User's basic resume information

Optional but recommended

  • MCP tools (corporate legal risk category):
    • get_executed_entity
    • get_dishonest_entity
    • get_consumption_restrictions
    • get_final_cases
    • get_judgments
  • Deep Research tools:
    • Metaso AI Search (deep research mode)
    • Perplexity Pro

Optional Extensions

  • Dify.ai (for interactive resume)
  • Live Server (VS Code extension, for local testing)

Key Rules

1. Tool-Level Degradation Principle

Each MCP tool determines availability independently; when unavailable, degrade individually to Web Search. Do not let a single tool failure affect the overall process.

2. Information Authenticity Principle

All generated content must be based on:

  • Real information provided by the user
  • Publicly verifiable company/law firm information
  • Results returned by MCP or Web Search

It is strictly prohibited to fabricate experience, exaggerate achievements, or invent company information.

3. Structured Expression First

Prioritize using:

  • Tables (comparison, classification, risk matrix)
  • ASCII flowcharts (business processes, decision paths)
  • Lists (talking points, to-do items)
  • Timelines (development stages, career trajectories)

4. Interview-Oriented Principle

All generated materials must be able to directly support the interview:

  • Every piece of information in the memo must be convertible into conversation content
  • Every experience in the resume must be prepared for elaboration
  • Every point in the risk analysis must have a response strategy

5. User Awareness Principle

Clearly inform the user:

  • Which information comes from MCP (high accuracy)
  • Which information comes from Web Search (requires verification)
  • Which information needs to be supplemented by the user (such as law firm private-domain information)

Quick Start (User Side)

Scenario 1: First-Time Use, Starting from Scratch

用户:我想申请腾讯音乐的法务岗位,JD在这里

Claude:
1. 提取 JD 信息
2. 询问:你有秘塔AI搜索或 Perplexity Pro 吗?
3. 开始调研:
   - 公司调研(Deep Research 或 Web Search)
   - 法律风险分析(MCP 或 Web Search)
4. 询问:你的基础简历信息
5. 生成:网页简历 + 面试备忘录

Scenario 2: Existing Resume, Adjusting Goals

用户:这是我的简历,我想申请红杉律所的资本市场团队

Claude:
1. 加载现有简历
2. 询问:你有关于这个团队的私域信息吗?
3. 使用「律所调研框架」+「团队信息收集清单」
4. 生成:针对性简历 + 面试备忘录

Scenario 3: Quick Research Mode

用户:帮我快速调研一下字节跳动的法律风险情况

Claude:
1. 直接执行法律风险分析(工具级降级)
2. 输出:法律风险分析报告

Extended Features

Interactive Resume Integration

See extensions/interactive-resume/dify-integration.md

Custom Resume Styles

See references/styles/custom-guide.md


References

Prompt Templates

  • references/prompts/preparation-checklist.md - Materials Preparation Checklist
  • references/prompts/html-resume-template.md - Resume HTML Generation Prompt
  • references/prompts/materials-prompt-template.md - Job Application Materials Generation Prompt
  • references/prompts/memo-word-spec.md - Interview Memo Word Document Specification
  • references/prompts/firm-research-extension.md - Law Firm Job Application Extension

Research Frameworks

  • references/frameworks/company-research-framework.md - Company Research Framework
  • references/frameworks/firm-research-framework.md - Law Firm Research Framework
  • references/frameworks/legal-risk-framework.md - Legal Risk Analysis Framework

MCP Tools

  • references/mcp-tools/mcp-catalog.md - Supported MCP Tools Catalog (Feature Matching)
  • references/mcp-tools/fallback-queries.md - MCP Fallback Query Template

Resume Styles

  • references/styles/business-elite.md - Business Elite Style Configuration
  • references/styles/modern-tech.md - Modern Tech Style Configuration
  • references/styles/minimal-clean.md - Minimalist Fresh Style Configuration
  • references/styles/custom-guide.md - Custom Configuration Guide