Skip to content

README.md

Latest commit

 

History

History
73 lines (54 loc) · 3.48 KB

README.md

File metadata and controls

73 lines (54 loc) · 3.48 KB
Raw
Strix AI: Run Your Own Autonomous AI Hacker Locally

📺 Watch the full tutorial on YouTube

🦉 Strix Visual Explorer

A lightweight Streamlit interface for Strix, the open-source AI pentesting tool. This project provides a visual way to run Strix security scans against local directories or web applications using local LLMs (via Ollama).

Strix Streamlit Ollama

🛠️ Tech Stack

  • Core Security Tool: Strix (Autonomous AI Pentesters)
  • UI Framework: Streamlit
  • LLM Engine: Ollama (qwen2.5-coder:1.5b, reasoning/think=false for speed)
  • Language: Python 3.10+

📁 File Explanations

  • app.py: Main Streamlit application providing the visual interface to configure and launch Strix scans.
  • requirements.txt: Minimal dependencies required to run the visual interface.
  • dummy_app/dummy_target.py: A simple vulnerable python script used as a target for demonstrating the scanning capabilities.
  • output.md: Exact output report generated by the Strix scan on the dummy target.

🚀 Setup & Install

Note: This repository contains the Streamlit visual interface. If you want to modify Strix's source code directly, you must clone their main repository: git clone https://github.com/usestrix/strix.git

Run these commands in your Windows PowerShell terminal:

# 1. Install dependencies
pip install -r requirements.txt

# 2. Install Strix (if not already installed, via pip)
pip install strix-agent

# 3. Ensure Ollama is running with qwen2.5-coder:1.5b
ollama pull qwen2.5-coder:1.5b

🎮 Run

# Launch the Streamlit application
streamlit run app.py

🧪 Test

# Verify Strix CLI is correctly configured
strix --help

💡 5 Use Cases

  1. Local Code Review: Quickly scan your local project directory for security vulnerabilities before committing code.
  2. Web App Assessment: Point the tool at a staging URL to identify OWASP Top 10 vulnerabilities.
  3. Continuous Security: Use the headless mode integrated via the UI's underlying commands for scheduled checks.
  4. Vulnerability Verification: Validate findings with AI-generated Proof-of-Concepts (PoCs) without manual exploit writing.
  5. Security Education: Learn about vulnerabilities and remediation strategies by scanning intentional vulnerable apps (like dummy_target.py).

🔮 5 Future Features

  1. Report Exporting: Add PDF and HTML export options for Strix scan results directly from the UI.
  2. Scan History Tracker: A built-in database to track vulnerabilities found over time across different targets.
  3. Agent Persona Selection: Allow selecting different Strix agent personas (e.g., Recon specialist, Exploitation expert) via the UI.
  4. Custom Prompt Injection: UI fields to provide custom testing instructions or scope rules to the Strix agents.
  5. CI/CD Configuration Generator: A button to automatically generate GitHub Actions or GitLab CI YAML files for the scanned repository.

Keywords: Strix AI, Autonomous Pentesting, Local LLM Security, Ollama Pentesting, Qwen Coder Security, Streamlit Cybersecurity, Automated Vulnerability Scanner, AI Hacking Tool.