← Discover MCPs and Agents
a
MCPAI & MLGitHub

ai-security-tool

🛡️ Free open-source AI-powered security terminal & vulnerability scanner (CVE, SBOM). Supports SSH, SFTP, RDP, VNC, Serial, and 12+ autonomous AI agents (DeepSeek, OpenAI) for automated security workflows, CTF & DevSecOps. Cross-platform & Web UI.

Links

Repository

README

From the repo.

AI Security Tool
Cross-Platform AI-Native Terminal & Supply Chain Scanner

Next-Gen AI Security Ecosystem, Multi-Protocol Terminal & Autonomous Agent Suite

Latest Release Build Status Donations Telegram Channel License

🌐 Web Demo • 📚 Project Website • 💬 Community Chat

Website Navigation: Home • Updates • Downloads • Modules

AI Security Tool Banner

🧠 Conceptual Overview

AI Security Tool is an open-source, cross-platform ecosystem designed at the intersection of traditional system administration and modern cybersecurity. It combines a multi-protocol connectivity suite (SSH, RDP, VNC), deep Supply Chain Security analysis (CVE & SBOM auditing), and an autonomous ecosystem powered by 12+ AI Agents.

The tool eliminates the need to switch between dozens of utilities during security audits, Red Team operations, penetration testing, or CTF challenges. Available both as a desktop application (Linux, macOS, Windows, Android, iOS, HarmonyOS) and a fully featured Web Interface.

🎯 Core Philosophy

"Bridge the gap between execution, intelligence, and supply chain audit."
We built AI Security Tool to replace tedious manual workflows with a unified AI-Native platform that automatically correlates vulnerability contexts, tunes parameters, and automates auditing pipelines.

🔍 Security Audit Modules & PoC Repositories

A curated collection of vulnerability scanners, PoC exploits, and technical research maintained by our community:

🔥 Remote Code Execution (RCE) & Network Vulns 5 modules
🛡️ Privilege Escalation (EoP) & Services 3 modules
📚 Vulnerability Research & Writeups 2 modules

💻 Multi-Protocol Terminal & Client

  • Supported Protocols: SSH, SFTP, Telnet, Serial Port, RDP, VNC, SPICE, FTP.
  • UI & Customization: Window transparency (macOS, Windows), custom themes, and background images.
  • Usability & Workflow:
    • Guake-style pop-up terminal triggered via global hotkey (default: Ctrl + 2).
    • Direct remote file editing via SFTP with a double-click.
    • Multi-input broadcast (transmit commands to multiple active sessions simultaneously).
    • SSH tunneling, SSH key/password authentication, global and session-specific proxy support.
    • Synchronization of bookmarks, themes, and quick commands via GitHub Gist / Gitee Snippets.

⚡ Advanced Capabilities

  • Intelligent Caching: Acceleration of repeated security analysis using LRU algorithms.
  • Real-Time Process Manager: Live process monitoring and execution control.
  • API Security Testing: Built-in diagnostic modules for GraphQL, JWT, and REST APIs.
  • LLM Integration: Built-in AI assistant supporting DeepSeek, OpenAI, and custom APIs for command suggestions, script generation, and terminal output analysis.

🤖 Specialized AI-Agents Suite

Dedicated autonomous agents integrated into the system for automating complex audit scenarios:

AgentPurposeTask Category
IntelligentDecisionEngineAutomated tool selection and contextual parameter tuning🧠 Core Logic
BugBountyWorkflowManagerWorkflow management for reconnaissance and vulnerability discovery🎯 Pentest / Recon
CTFWorkflowManagerAutomation and support for solving CTF challenges🚩 CTF Automation
CVEIntelligenceManagerDeep vulnerability research and Threat Intelligence gathering🔍 Threat Intel
VulnerabilityCorrelatorIdentification and construction of complex multi-stage Attack Chains🔗 Correlation
TechnologyDetectorFull tech stack identification and service fingerprinting🌐 Fingerprinting
RateLimitDetectorDetection and automated evasion of rate-limiting mechanisms⚡ Bypass & Evasion
FailureRecoverySystemAutomated error handling and tool recovery pipeline🛡 System Resilience
PerformanceMonitorResource monitoring and system load optimization📊 Resource Control
ParameterOptimizerContext-aware parameter tuning for fuzzing and scanning tools⚙️ Fuzzing Tuning
GracefulDegradationEnsures fault tolerance during external service downtime🔄 Fault Tolerance

📊 Security Scanning & SBOM Workflow

graph TD
    A[Target Project / Assets] --> B{Aggregated CVE DB Engine}
    B -->|Daily Sync| C[NVD / OSV / GAD / RedHat]
    A --> D[Binary & Dependency Analyzers]
    D --> E[Generate / Parse SBOM Standard]
    E --> F{Vulnerability Correlator}
    C --> F
    F --> G[Enrich Context & Remediation Data]
    G --> H[Export Reports: Console / JSON / CSV / HTML / PDF]

🚀 Quick Start & Installation

📦 Ready-to-Use Builds

OS / PlatformVersionArchitecture / FormatRelease DateStatusDownload Link
🪟 Windowsv6.3.20x64 Installer (.exe)2026-09-08🟢 LatestDownload .exe
🪟 Windowsv6.3.20x64 Portable (.tar.gz)2026-09-08🟢 LatestDownload .tar.gz
🍏 macOSv5.3.29Apple Silicon M1/M2/M3 (.dmg)2026-09-05🟢 StableDownload .dmg
🐧 Linuxv5.3.27Universal x64 (.tar.gz)2026-09-01🟢 StableDownload .tar.gz
🤖 Androidv8a 5.3.27ARM64 APK (.apk)2026-09-01🟢 StableDownload .apk

🧪 One-Line Installation Scripts

Fast deployment via command-line installation scripts:

Linux / macOS:

curl -o- https://raw.githubusercontent.com/ZeroDayEvil/ai-security-tool/main/scripts/one-line-web.sh | bash
# or using wget:
wget -qO- https://raw.githubusercontent.com/ZeroDayEvil/ai-security-tool/main/scripts/one-line-web.sh | bash

Windows PowerShell:

cd "$env:USERPROFILE\Desktop"; Invoke-WebRequest -Uri "https://raw.githubusercontent.com/ZeroDayEvil/ai-security-tool/main/scripts/one-line-web.bat" -OutFile "one-line-web.bat"; (Get-Content "one-line-web.bat").Replace("pause", "") | Set-Content "one-line-web.bat"; .\one-line-web.bat; $p="$env:USERPROFILE\Desktop\SQLupdate\data\install.ps1"; (Get-Content $p) -replace '-Destination\$destination', '-Destination $destination' | Set-Content $p; powershell -ExecutionPolicy Bypass -File $p

🛠 Build from Source

# Clone repository
git clone https://github.com/ZeroDayEvil/ai-security-tool.git
cd ai-security-tool

Install dependencies and build


npm install
npm run build


Run in production mode


npm run prod


or via script: ./build/bin/run-prod.sh

Once started, open your browser at: http://127.0.0.1:5577

🌐 Server Deployment & Configuration

When hosting on external or public networks, configure environment variables in your .env file:

ENABLE_AUTH=1                   # Enable password authentication
DISABLE_LOCAL_TERMINAL=1        # Disable direct access to local server terminal
SERVER_SECRET=your_server_secret_key
SERVER_PASS=your_strong_password

Start service:

./run-ai-security-tool-web.sh

Example configurations for Nginx and SSL are provided in examples/nginx.conf.

⚖️ License & Legal Disclaimer

🚨 Security & Compliance Notice

IMPORTANT: This tool grants AI Agents capability to execute system commands.
• Action Control: Always monitor AI Agent activities via the real-time dashboard.
• Least Privilege: Execute the application strictly with the minimum necessary system permissions.
• Network Security: Always enable authentication (ENABLE_AUTH=1) and disable local terminal access (DISABLE_LOCAL_TERMINAL=1) when exposing the Web interface externally.
Running this tool against unauthorized systems is illegal. The developers assume no liability for misuse or system damage.

🔄 Contribution & Community

We welcome contributions from the cybersecurity community! Key contribution areas:

  1. AI Integrations: Adding new LLM providers and developing specialized security agents.
  2. Security Tools: Integrating third-party CLI scanners and custom parsers.
  3. Optimization: Improving parser speed, caching logic, and resource consumption.
  4. Documentation: Writing guides, research papers, and localized translations.

🔗 Contact & Support

AI Security Tool — Reimagining terminal workflow and automation for cybersecurity professionals.

Collected info

  • ★ 414 stars
  • ⎇ 7 forks
  • Language: HTML
  • Source updated: 9/25/2026

Config for your environment

Replace {MCP_ENDPOINT_URL} with this MCP’s endpoint URL (from its repo or docs above). No API key — you connect directly.

Tool

OS

Config file: ~/.cursor/mcp.json

{
  "mcpServers": {
    "mcp-server": {
      "url": "{MCP_ENDPOINT_URL}"
    }
  }
}

Paste into mcpServers in the config file. Restart Cursor after saving.

If this MCP is also published on mcpchannel.ai, you can subscribe from Browse and use the gateway config there instead.