BitSight Community MCP Server
Enables access to BitSight security rating data through company search and rating retrieval functionality. Automatically manages BitSight subscriptions for one-time rating requests and provides structured security assessments for companies.
README
<div align="center"> <img src="birre-logo.png" alt="Birre Logo" width="375"> </div>
A Model Context Protocol (MCP) server that provides access to BitSight security rating data. It utilizes FastMCP and the brandon-smith-187/bitsight Python library. It can be run easily without installation in a temporary, isolated Python environment with uv.
Installation
Configuration
Set these environment variables:
BST_API_KEY(required): Your BitSight API keyDEBUG(optional): Enable debug logging (true,false, default:false)
export BST_API_KEY="your-bitsight-api-key"
Run directly from GitHub with uvx
uvx --from git+https://github.com/boecht/birre server.py
Or run locally
git clone https://github.com/boecht/birre
uv run server.py
That's it! The script will automatically install all dependencies using PEP 723 inline metadata.
Alternatively run with fastmcp for more options, like HTTP transport.
Disclaimer
This project is not affiliated with, endorsed by, or sponsored by BitSight Technologies, Inc. This is an unofficial, community-developed MCP server that provides integration with Bitsight's publicly available services.
- This project is developed and maintained independently by the open source community
- "Bitsight" is a registered trademark of BitSight Technologies, Inc.
- This integration is provided "as-is" without any warranty or official support from BitSight Technologies, Inc.
This project enables third-party access to Bitsight services through their public APIs and is intended for educational and integration purposes only.
Features
Version 1.0 (MVP)
- Company Search: Search for companies by name or domain
- Company Rating: Get security ratings with automatic subscription management
- Ephemeral Subscriptions: Automatically subscribe and unsubscribe for one-time rating requests
- Structured Error Handling: Clear error responses for API issues and quota limits
- uv/uvx Compatible: Run easily with uv using PEP 723 inline script metadata
Available Tools
company_search
Search for companies in the BitSight database.
Parameters:
name(optional): Company name to search fordomain(optional): Company domain to search for
Returns:
companies: List of matching companies withGUID,name,domain, andindustrycount: Total number of matchessearch_term: The term used for searching
Example:
{
"companies": [
{
"guid": "e90b389b-0b7e-4722-9411-97d81c8e2bc6",
"name": "GitHub, Inc.",
"domain": "github.com",
"industry": "Technology"
},
...
],
"count": 4,
"search_term": "GitHub"
}
get_company_rating
Get security rating for a company by GUID. Automatically manages BitSight subscriptions.
Parameters:
guid(required): BitSight GUID of the company
Returns:
rating: Current security rating (0-900 scale)rating_date: Date of the ratinggrade: Letter grade (A, B, C, D, F)company_name: Name of the companyauto_subscribed: Boolean indicating if auto-subscription was performedauto_unsubscribed: Boolean indicating if auto-unsubscription was performed
Example:
{
"rating": 740,
"rating_date": "2025-08-26",
"grade": "A",
"company_name": "GitHub, Inc.",
"auto_subscribed": true,
"auto_unsubscribed": true
}
Development
Project Structure
birre/
├── server.py # PEP 723 entry point script
├── src/
│ └── birre/
│ ├── __init__.py
│ ├── server.py # Main FastMCP server
│ ├── config.py # Configuration management
│ └── tools/
│ ├── __init__.py
│ ├── company_search.py
│ └── company_rating.py
├── tests/
│ ├── unit/ # Unit tests with mocked dependencies
│ └── integration/ # Integration tests with real API calls
└── requirements.txt # Development dependencies
Future Versions
Version 2.0: Category Risk Ratings
- Retrieve risk ratings by category/vector
- Filter by specific risk categories
Version 3.0: Company Reports
- Download official BitSight PDF reports
- Handle report generation limits
Version 4.0: Database Caching
- Daily caching of company rating data
- Reduce duplicate API calls
Version 5.0: Multi-Tenant Service
- Remote deployment support
- Authentication and authorization
- Concurrent user support
License
This project is licensed under the MIT License.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。