Spaces:
Sleeping
Sleeping
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,12 +1,64 @@
|
|
| 1 |
---
|
| 2 |
title: Sentinel MLOps Agent
|
| 3 |
-
emoji:
|
| 4 |
-
colorFrom:
|
| 5 |
-
colorTo:
|
| 6 |
sdk: docker
|
|
|
|
| 7 |
pinned: false
|
| 8 |
-
license: mit
|
| 9 |
-
short_description: Autonomous Anomaly Detection & RAG Investigation Agent
|
| 10 |
---
|
| 11 |
|
| 12 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
title: Sentinel MLOps Agent
|
| 3 |
+
emoji: 🛡️
|
| 4 |
+
colorFrom: blue
|
| 5 |
+
colorTo: indigo
|
| 6 |
sdk: docker
|
| 7 |
+
app_port: 7860
|
| 8 |
pinned: false
|
|
|
|
|
|
|
| 9 |
---
|
| 10 |
|
| 11 |
+
<div align="center">
|
| 12 |
+
|
| 13 |
+
# 🛡️ Sentinel: Autonomous MLOps Agent
|
| 14 |
+
### *Self-Healing Infrastructure Powered by Physics & Gemini 2.5*
|
| 15 |
+
|
| 16 |
+
[](https://fastapi.tiangolo.com/)
|
| 17 |
+
[](https://ai.google.dev/)
|
| 18 |
+
[]()
|
| 19 |
+
|
| 20 |
+
</div>
|
| 21 |
+
|
| 22 |
+
---
|
| 23 |
+
|
| 24 |
+
## ⚡ What is Sentinel?
|
| 25 |
+
Sentinel is an autonomous monitoring agent that doesn't just alert you when a server crashes—it **investigates why.**
|
| 26 |
+
|
| 27 |
+
Most MLOps tools rely on static thresholds (*"Alert if CPU > 80%"*). Sentinel uses **Statistical Physics (Z-Scores)** to learn the "normal" baseline of your system and detect subtle **Data Drift**.
|
| 28 |
+
|
| 29 |
+
When an anomaly is detected, it triggers an AI Agent (**Gemini 2.5 Flash**) to perform **Root Cause Analysis (RCA)** on the system logs automatically.
|
| 30 |
+
|
| 31 |
+
---
|
| 32 |
+
|
| 33 |
+
## 🚀 Live Demo: How to Test It
|
| 34 |
+
This API is live. You can simulate a "Cyber Attack" right now.
|
| 35 |
+
|
| 36 |
+
1. **[Click Here to Open the Swagger UI](/docs)** (or scroll down if embedded).
|
| 37 |
+
2. Click the green **`POST /monitor`** bar.
|
| 38 |
+
3. Click **Try it out**.
|
| 39 |
+
4. **The Test:** Change `cpu_usage` to **900**.
|
| 40 |
+
```json
|
| 41 |
+
{
|
| 42 |
+
"timestamp": "12:00",
|
| 43 |
+
"service_name": "PaymentGateway",
|
| 44 |
+
"cpu_usage": 900
|
| 45 |
+
}
|
| 46 |
+
```
|
| 47 |
+
5. Click **Execute**.
|
| 48 |
+
|
| 49 |
+
### 📉 The Result
|
| 50 |
+
You will receive a **CRITICAL** alert containing a full AI investigation report:
|
| 51 |
+
> *"Root Cause: Unauthorized crypto-mining process detected. Recommended Fix: Terminate process PID 409..."*
|
| 52 |
+
|
| 53 |
+
---
|
| 54 |
+
|
| 55 |
+
## ⚙️ Architecture
|
| 56 |
+
|
| 57 |
+
```mermaid
|
| 58 |
+
graph LR
|
| 59 |
+
A[Live Data Stream] --> B(FastAPI Endpoint)
|
| 60 |
+
B --> C{Anomaly Detector (Z-Score)}
|
| 61 |
+
C -- Normal --> D[Log Metric]
|
| 62 |
+
C -- "Anomaly (>2.5σ)" --> E[Gemini 2.5 RAG Agent]
|
| 63 |
+
E --> F[Retrieve Error Logs]
|
| 64 |
+
F --> G[Generate Incident Report]
|