mithra916/IAMpact--Security-Log-Analyzer
GitHub: mithra916/IAMpact--Security-Log-Analyzer
IAMpact 是一款 AI 驱动的 AWS CloudTrail IAM 日志分析器,通过交互式 SOC 仪表板检测可疑身份活动并提供风险评分和修复建议。
Stars: 0 | Forks: 0
# IAMpact — 安全日志分析器
IAMpact 是一个轻量级的 AI 驱动的 IAM 安全分析仪表板,可帮助安全团队分析 AWS CloudTrail IAM 日志、检测可疑的身份活动、确定警报优先级并生成修复建议。
当前版本作为一个交互式的 **CloudTrail 日志分析器**运行,用户可以粘贴 AWS IAM 日志并立即接收经过风险评分的警报。
## 当前功能
- 直接将 AWS CloudTrail IAM 日志粘贴到仪表板中
- 解析并规范化 IAM 事件
- 检测敏感的 IAM 操作
- 为每个事件分配风险评分
- 将警报分类为 LOW、MEDIUM、HIGH 或 CRITICAL
- 高亮显示最高风险的警报
- 显示检测原因
- 生成 AI 风格的修复建议
- 可视化的 SOC 风格仪表板 UI
- 采用 FastAPI 后端与 React 前端
## 工作原理
```
CloudTrail Logs
↓
Paste into Dashboard
↓
FastAPI Analyzer API
↓
Log Parsing & Normalization
↓
Rule-Based Risk Scoring
↓
Alert Prioritization
↓
AI Recommendation Engine
↓
SOC Dashboard Visualization
```
检测示例
```
Input CloudTrail event:
{
"eventTime": "2026-05-26T10:20:00Z",
"eventName": "CreateAccessKey",
"sourceIPAddress": "45.67.89.10",
"awsRegion": "us-east-1",
"userIdentity": {
"userName": "admin"
}
}
```
输出:
Priority: HIGH
Risk Score: 65
User: admin
Action: CreateAccessKey
Source IP: 45.67.89.10
Reasons:
- 检测到敏感的 IAM 操作
- 检测到外部源 IP
Recommendations:
- 调查用户会话
- 验证该活动是否已获授权
- 监控额外的 IAM 活动
## 项目结构
```
cloud_sec_intel/
│
├── backend/
│ ├── app/
│ │ ├── ai/
│ │ ├── api/
│ │ ├── core/
│ │ ├── detection/
│ │ ├── ingestion/
│ │ ├── models/
│ │ ├── parsing/
│ │ ├── schemas/
│ │ ├── services/
│ │ ├── threat_intel/
│ │ ├── websocket/
│ │ └── main.py
│ └── requirements.txt
│
├── frontend/
│ ├── src/
│ │ ├── pages/
│ │ │ └── Dashboard.jsx
│ │ ├── App.jsx
│ │ └── main.jsx
│ └── package.json
│
├── docker-compose.yml
├── .gitignore
└── README.md
```
## 技术栈
```
Backend
Python
FastAPI
Uvicorn
Rule-based risk scoring
IAM event parsing
AI-style recommendation engine
Frontend
React
Vite
JavaScript
```
自定义 SOC 仪表板 UI
▶ 如何在本地运行
```
1. Clone the Repository
git clone https://github.com/mithra916/IAMpact.git
cd IAMpact
2. Run Backend
cd backend
python -m venv venv
venv\Scripts\activate
pip install -r requirements.txt
uvicorn app.main:app --reload
Backend will run at:
http://127.0.0.1:8000
API docs:
http://127.0.0.1:8000/docs
3. Run Frontend
Open new terminal:
cd frontend
npm install
npm run dev
Frontend will run at:
http://localhost:5173
```
## 测试日志
将此内容粘贴到仪表板中:
```
[
{
"eventTime": "2026-05-26T10:20:00Z",
"eventName": "CreateAccessKey",
"sourceIPAddress": "45.67.89.10",
"awsRegion": "us-east-1",
"userIdentity": {
"userName": "admin"
}
},
{
"eventTime": "2026-05-26T10:25:00Z",
"eventName": "DeleteTrail",
"sourceIPAddress": "185.220.101.45",
"awsRegion": "ap-south-1",
"userIdentity": {
"userName": "cloud-admin"
}
},
{
"eventTime": "2026-05-26T10:30:00Z",
"eventName": "ListUsers",
"sourceIPAddress": "192.168.1.20",
"awsRegion": "ap-south-1",
"userIdentity": {
"userName": "auditor"
}
}
]
```
## 当前仪表板面板
```
Total logs analyzed
Alerts detected
Highest risk score
Top priority level
Severity breakdown
Risk score trend
IAM action breakdown
Top security alert
Detected alerts list
AI recommendations
```
## 检测逻辑
```
IAMpact currently uses rule-based security scoring.
Risk factors include:
Sensitive IAM actions
External source IPs
Failed authentication attempts
CloudTrail tampering events
High-risk AWS API calls
Example high-risk actions:
CreateAccessKey
DeleteAccessKey
AttachUserPolicy
AttachRolePolicy
CreateUser
DeleteUser
DeactivateMFADevice
DeleteTrail
StopLogging
AssumeRole
ConsoleLogin
```
### 路线图
计划中的未来升级:
```
Real AWS CloudTrail integration
STS AssumeRole-based AWS account connection
Multi-account monitoring
WebSocket live alert streaming
Threat intelligence enrichment using AbuseIPDB
GeoIP visualization
ML-based anomaly detection
PostgreSQL alert storage
SOC analyst case management
Slack or email notifications
```
## 项目状态
当前状态:
```
✅ Working CloudTrail Log Analyzer
✅ FastAPI backend
✅ React SOC dashboard
✅ Risk scoring engine
✅ AI recommendation output
🚧 Real-time AWS integration planned
🚧 Threat intelligence enrichment planned
🚧 ML anomaly detection planned
```
```
Author
Loga Mithra R
Cybersecurity Student | AI & Security Automation
```
标签:AI安全分析, AMSI绕过, AV绕过, FastAPI, IAM安全, React, Syscalls, 威胁检测, 安全运营, 扫描框架, 数据可视化, 请求拦截, 逆向工具