AWS Security Analyzer
An MCP server for automated AWS security scanning. It detects over-privileged IAM roles and S3 Public Access Block misconfigurations, enabling least-privilege enforcement through tools like list_iam_roles, analyze_iam_policy, and analyze_s3_security.
README
AWS Security Analyzer (MCP Server)
An automated Cloud Security Posture Management (CSPM) tool built on the Model Context Protocol (MCP). This server acts as a local security scanner for AWS environments, designed to detect over-privileged IAM roles and S3 Public Access Block misconfigurations using the Principle of Least Privilege.
🚀 Features
- IAM Role Enumeration: Lists IAM roles and ARNs with pagination support.
- Policy Analysis: Scans both managed and inline IAM policies for
AdministratorAccessand wildcard (*) action/resource risks. - S3 Public Access Block Scanner: Audits S3 buckets for missing or partial Public Access Block (PAB) settings.
🛠️ Prerequisites
- Python 3.10+
- Node.js (optional, for MCP Inspector testing)
- AWS credentials configured (
aws configure, environment variables, or an IAM role)
⚙️ Installation
- Clone the repository:
git clone https://github.com/ekremcakir/aws-security-mcp.git
cd aws-security-mcp
- Create and activate a virtual environment:
python3 -m venv .venv
macOS / Linux:
source .venv/bin/activate
Windows:
.venv\Scripts\activate
- Install dependencies:
pip install -r requirements.txt
🔍 Usage
MCP Inspector (local testing)
npx @modelcontextprotocol/inspector .venv/bin/python server.py
Open the localhost URL in your browser, go to Tools, and run:
list_iam_rolesanalyze_iam_policyanalyze_s3_security
Cursor / Claude Desktop
Add this to your MCP config (adjust paths to your machine):
{
"mcpServers": {
"aws-security": {
"command": "/absolute/path/to/aws-security-mcp/.venv/bin/python",
"args": ["/absolute/path/to/aws-security-mcp/server.py"]
}
}
}
Windows example:
{
"mcpServers": {
"aws-security": {
"command": "C:\\path\\to\\aws-security-mcp\\.venv\\Scripts\\python.exe",
"args": ["C:\\path\\to\\aws-security-mcp\\server.py"]
}
}
}
🔐 Required AWS IAM Permissions
The IAM user or role running this server needs at least:
| Service | Actions |
|---|---|
| IAM | iam:ListRoles, iam:ListAttachedRolePolicies, iam:ListRolePolicies, iam:GetRolePolicy, iam:GetPolicy, iam:GetPolicyVersion |
| S3 | s3:ListAllMyBuckets, s3:GetPublicAccessBlock |
Example read-only policy snippet:
{
"Version": "2012-10-17",
"Statement": [
{
"Effect": "Allow",
"Action": [
"iam:ListRoles",
"iam:ListAttachedRolePolicies",
"iam:ListRolePolicies",
"iam:GetRolePolicy",
"iam:GetPolicy",
"iam:GetPolicyVersion",
"s3:ListAllMyBuckets",
"s3:GetPublicAccessBlock"
],
"Resource": "*"
}
]
}
⚠️ Limitations
- S3 scope: Checks Public Access Block settings only. Does not analyze bucket policies, ACLs, or actual public object exposure.
- IAM scope: Does not evaluate permission boundaries, trust policies, or cross-account access patterns.
- Pagination:
list_iam_rolesrespectsmax_items; large accounts may need multiple calls with higher limits. - Credentials: Uses the default Boto3 credential chain (environment, shared config, instance profile).
🛡️ Architecture & DevSecOps Context
This project demonstrates cloud security automation with standard AWS SDKs (Boto3) and the MCP standard, bridging infrastructure auditing and AI-driven workflow integrations.
📄 License
MIT — see 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。