Skip to main content
Glama
justingyang218

劳动&劳务合同风险排查 MCP Server

Labor & Service Contract Risk Review MCP Server

A contract risk review MCP service built on FastMCP that can be invoked by any MCP protocol-compatible client (DeepSeek Harness, Claude Code, Cursor, Windsurf, etc.).

✨ Features

  • Single Contract Review: Conducts comprehensive intelligent review of labor contracts (16 risk categories) or service contracts (9 risk categories)

  • Batch Contract Review: Supports batch review of multiple contracts and automatically generates an overview table

  • Risk Knowledge Base Query: Queries key risk review points and legal basis for a specified category

  • Local Regulation Query: Queries local regulation differences for Beijing/Shanghai/Guangdong/Zhejiang/Jiangsu

  • Automatic Contract Type Recognition: Intelligently determines whether a contract is a labor contract or a service contract

  • Personal Information Desensitization: Automatically desensitizes ID card numbers, phone numbers, and bank card numbers

Related MCP server: Contract Review AI MCP

📦 Installation

pip install -r requirements.txt

⚙️ Configuration

Environment Variables

Variable Name

Required

Default Value

Description

LLM_API_KEY

✅ Yes

-

LLM API key

LLM_BASE_URL

❌ No

https://api.openai.com/v1

API endpoint URL (OpenAI-compatible interface)

LLM_MODEL

❌ No

gpt-4o-mini

Model name

LLM_TEMPERATURE

❌ No

0.1

Sampling temperature

Configuration Example

export LLM_API_KEY="sk-xxxxxxxxxxxxxx"
export LLM_BASE_URL="https://api.deepseek.com/v1"
export LLM_MODEL="deepseek-chat"

💡 Supports any OpenAI-compatible API interface (DeepSeek, Qwen, Zhipu AI, Ollama, etc.)

🚀 Running

Method 1: STDIO Mode (Local Tools/Desktop Clients)

python server.py
# 或显式指定
python server.py stdio

Method 2: HTTP Mode (Network Service)

python server.py http 0.0.0.0 8000

After startup, access http://localhost:8000/mcp to use it.

Method 3: Using the fastmcp CLI

# 开发调试
fastmcp dev server.py

# 安装到 Claude Desktop
fastmcp install server.py

🔌 MCP Tools

1. review_contract

Performs a complete risk review of the contract text.

Parameters:

  • contract_text (string, required): Contract text content

  • contract_type (string, optional): Contract type, auto/labor/service, default auto

Example:

review_contract(contract_text="劳动合同文本...", contract_type="auto")

2. batch_review

Reviews multiple contracts in batch.

Parameters:

  • contracts_text (string, required): Text of multiple contracts

  • separator (string, optional): Separator, default ---合同分割线---

Example:

batch_review(contracts_text="合同1\n---合同分割线---\n合同2", separator="---合同分割线---")

3. get_risk_knowledge

Queries the risk knowledge base for a specified category.

Parameters:

  • category (string, required): Risk category, e.g., "probation period", "non-compete restriction", "personal injury compensation"

Example:

get_risk_knowledge(category="竞业限制")

4. get_local_regulations

Queries regulation differences for a specified region.

Parameters:

  • region (string, required): Region, Beijing/Shanghai/Guangdong/Zhejiang/Jiangsu

  • topic (string, required): Topic, e.g., non-compete compensation/overtime pay base/minimum wage/maternity leave/marriage leave

Example:

get_local_regulations(region="上海", topic="竞业补偿")

📚 MCP Resources

URI

Description

risk-knowledge://categories

List of all risk categories

risk-knowledge://labor

List of 16 risk categories for labor relations

risk-knowledge://service

List of 9 risk categories for service relations

local-regions://list

List of supported regions

💬 MCP Prompts

Name

Description

contract_review_prompt

Complete contract review prompt template

batch_review_prompt

Batch review prompt template

🔗 Client Integration

Claude Desktop

Add the following to claude_desktop_config.json:

{
  "mcpServers": {
    "contract-risk-review": {
      "command": "python",
      "args": ["/path/to/server.py"],
      "env": {
        "LLM_API_KEY": "your-api-key",
        "LLM_BASE_URL": "https://api.openai.com/v1",
        "LLM_MODEL": "gpt-4o-mini"
      }
    }
  }
}

Cursor / Windsurf

Simply add the above stdio configuration to the IDE's MCP configuration.

HTTP/SSE Clients

Set the MCP server address to http://localhost:8000/mcp.

📁 Directory Structure

contract-risk-review-mcp/
├── server.py              # MCP Server 主文件
├── requirements.txt       # 依赖清单
├── README.md              # 使用说明
└── knowledge/             # 知识库
    ├── risk-knowledge-base.md    # 劳动关系16类风险知识库
    ├── service-contract-risks.md # 劳务关系9类风险知识库
    ├── local-regulations.md      # 地方法规差异知识库
    └── categories-index.md       # 风险类别索引

⚠️ Disclaimer

This tool only provides contract risk references, clause modification suggestions, and regulatory basis. It does not constitute formal legal advice and cannot replace the professional review of a licensed attorney. For matters involving labor arbitration or major rights disputes, be sure to consult a licensed attorney.

📝 Changelog

v1.0.0

  • Initial version

  • Supports review of 16 risk categories for labor contracts

  • Supports review of 9 risk categories for service contracts

  • Supports batch review, knowledge base queries, and local regulation queries

  • Supports both STDIO and HTTP/SSE transport modes

F
license - not found
Not graded
quality - not tested
C
maintenance

Maintenance

Maintainers
Response time
Release cycle
Releases (12mo)
Commit activity

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Servers

View all related MCP servers

Related MCP Connectors

  • ContractOracle - 10 contract analysis tools: clause extraction, redlines, DORA mappings.

  • Dispatch litigation work to legal-services vendors from any MCP-compatible AI workflow.

  • Hosted MCP for denial, prior auth, reimbursement, workflow validation, batch scoring, and feedback.

View all MCP Connectors

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/justingyang218/contract-risk-review-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server