zero-log - GitLite

KO | EN

gitlite — search

#css #nextjs #vue #android #llm #markdown #tailwindcss #hacktoberfest #gpt #svg

zero-log

(★ 9)

Client-side AI log analysis engine. Detect volumetric brute-force threats and anomalous UA patterns locally, with zero server-side data ingestion.

파일 탐색기

ZIP 다운로드

- favicon.svg
- icons.svg
- - hero.png
  - react.svg
  - vite.svg
- App.css
- App.jsx
- index.css
- main.jsx
.gitignore
eslint.config.js
generate_logs.py
index.html
malicious_payload.csv
package-lock.json
package.json
postcss.config.js
README.md
tailwind.config.js
vite.config.js

# CDN으로 사용하기

jsDelivr

jsDelivr는 공개 GitHub 리포지토리를 별도 설정 없이 CDN으로 즉시 서빙합니다. 버전과 파일을 고르면 웹페이지에 바로 붙일 수 있는 링크와 예시 코드가 만들어집니다.

버전

파일

링크

예시

# 프로젝트 배지

React
Vite
Tailwind CSS
Privacy
License: MIT

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

**ZERO-LOG** is a professional-grade, privacy-first threat hunting and log analysis engine engineered in React. It bridges the gap between rapid security auditing and strict data compliance by processing high-volume server logs (Nginx, Apache, AWS) entirely within the browser's local memory—ensuring zero data exfiltration.

🌐 **[Launch Live Dashboard](https://GiTaDi-CrEaTe.github.io/zero-log/)**

---

## 🧠 Core Intelligence & Architecture

Zero-Log goes beyond simple request counting. It utilizes a multi-layered heuristic engine written in JavaScript to identify malicious intent and deterministic attack patterns.

* **Zero-Trust Ingestion:** Utilizing the `PapaParse` library, the engine processes thousands of CSV log entries locally within the browser's sandbox, providing high-speed analysis with zero server latency or data leakage.
* **Volumetric Analysis:** The algorithm dynamically flags source IPs exceeding 50 requests (High Risk) or 200 requests (Critical Risk) to instantly identify Brute-Force and DDoS attack vectors.
* **UA Spoofing & Fingerprinting:** A built-in AI heuristic layer detects automated tooling signatures (e.g., `curl`, `python-requests`, `zgrab`, `wget`) even when attackers attempt to blend in with standard background traffic.
* **Deterministic Scoring:** Targets are assigned a risk score (0-100) based on endpoint targeting variety, frequency, and tool-based signatures, culminating in exportable WAF-ready JSON blocklists.

---

## 🚀 Execution & Simulation
```bash
### 1. Generate Simulated Attack Data
This repository includes a Python simulation script that generates a realistic server log (`malicious_payload.csv`) containing standard user traffic, a high-frequency Python botnet, and a critical-level endpoint brute-force attack.
# Run the simulation script to generate the CSV dataset
python generate_logs.py

2. Launch the React Engine

Start the local Vite development server to initialize the dashboard.
Bash

npm install
npm run dev

3. Hunt Threats

Open the local interface (http://localhost:5173).

Drag and drop the generated malicious_payload.csv into the Ingest Logs zone.

The engine will instantly parse the data, map the traffic distribution via Recharts, and isolate the threat actors.

Click Export JSON Dossier to download a structured threat report designed for immediate firewall integration.
⚙️ Tech Stack

Frontend Framework: React 18, Vite

Styling: Tailwind CSS

Data Processing: PapaParse (Client-side CSV Ingestion)

Data Visualization: Recharts

Icons: Lucide React