ZeroDayEvil/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.
📌 What is AI Security Tool?
A cross‑platform, open‑source suite that mixes a multi‑protocol terminal (SSH, RDP, VNC, etc.) with a supply‑chain & vulnerability‑audit engine. The heart of the product is a collection of 12+ autonomous AI agents that help security professionals automate reconnaissance, vulnerability correlation, and remediation‑report generation.
🎯 Core purpose
- Replace a toolbox of separate utilities (nmap, ssh clients, SBOM generators, CVE scanners) with a single UI that can:
- Connect to remote hosts via many protocols.
- Pull CVE data from NVD, OSV, RedHat, etc., and merge it with locally generated SBOMs.
- Run AI‑driven agents that suggest commands, write scripts, and stitch together multi‑stage attack chains.
- Target users: red‑teamers, bug‑bounty hunters, CTF participants, and any security analyst who wants a “terminal‑plus‑AI” workflow.
🛠️ Main features (as described in the README)
| Feature | What it does |
|---|---|
| Multi‑protocol terminal | SSH, SFTP, Telnet, Serial, RDP, VNC, SPICE, FTP. Pop‑up Guake‑style window, global hot‑key, theme/custom background, multi‑session broadcast, bookmark sync via Gist. |
| Supply‑chain scanner | Generates/parses SBOMs, syncs daily CVE databases, correlates findings, exports reports (JSON, CSV, HTML, PDF). |
| LLM integration | Built‑in assistant that can call DeepSeek, OpenAI or custom endpoints for command suggestions, script generation, and output analysis. |
| AI‑Agent suite | 12 named agents (e.g., IntelligentDecisionEngine, BugBountyWorkflowManager, VulnerabilityCorrelator) that automate tool selection, recon workflows, attack‑chain building, rate‑limit evasion, and error recovery. |
| Advanced utilities | Intelligent LRU caching, real‑time process manager, API security testing (GraphQL, JWT, REST), performance monitoring, parameter optimisation for fuzzers. |
| Cross‑platform builds | Native installers for Windows, macOS (Apple‑silicon), Linux, Android, and iOS. One‑line install scripts for Linux/macOS (bash) and Windows (PowerShell). |
| Web interface | Runs a local server (http://127.0.0.1:5577) that mirrors the desktop UI; can be deployed behind Nginx with optional auth. |
🤖 AI agents – quick glance
| Agent | Role |
|---|---|
IntelligentDecisionEngine |
Picks the right tool & tunes parameters automatically. |
BugBountyWorkflowManager |
Guides recon → vuln discovery → reporting for bug‑bounty programs. |
CTFWorkflowManager |
Automates common CTF tasks (enumeration, exploit chaining). |
CVEIntelligenceManager |
Pulls latest CVE intel, enriches with context, suggests mitigations. |
VulnerabilityCorrelator |
Links separate findings into multi‑stage attack chains. |
TechnologyDetector |
Fingerprints remote tech stacks. |
RateLimitDetector |
Detects and evades rate‑limiting. |
FailureRecoverySystem |
Handles tool crashes and retries. |
PerformanceMonitor |
Tracks CPU/memory usage, throttles when needed. |
ParameterOptimizer |
Tunes fuzzing/scanning parameters on the fly. |
GracefulDegradation |
Keeps the system functional if external services go down. |
... (others) |
Provide resilience, resource control, and fault tolerance. |
📦 Getting started (as per README)
- Download the appropriate installer for your OS (Windows
.exe, macOS.dmg, Linux.tar.gz, Android.apk). - One‑line install (Linux/macOS):
Windows PowerShell equivalent is provided.curl -o- https://raw.githubusercontent.com/ZeroDayEvil/ai-security-tool/main/scripts/one-line-web.sh | bash - Run the binary; a browser window opens at
http://127.0.0.1:5577. - (Optional) Edit a
.envfile to enable authentication, disable the local terminal, and set a secret before exposing the web UI publicly.
⚠️ Safety & legal notice
- The AI agents can execute system commands, so the README stresses monitoring via the dashboard, least‑privilege execution, and enabling auth when the web UI is reachable from outside.
- Using the tool against systems you don’t own is illegal; the authors disavow liability.
🤝 Community & contribution
- Contributions are welcomed for new LLM providers, additional security tools, performance tweaks, and documentation.
- Communication channels: Telegram chat, Open Collective for donations, and a web demo site.
TL;DR
AI Security Tool is a genuine, open‑source security platform that blends a rich remote‑access terminal with AI‑driven automation for vulnerability discovery, supply‑chain auditing, and reporting. It ships with dozens of pre‑built AI agents, LLM integration, and cross‑platform installers, aiming to be a one‑stop shop for modern penetration‑testing and bug‑bounty workflows.
Related
- Project
- Project
- Project
- Project
- Project