Skip to content

projectboot/drivershield

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 

Repository files navigation

DriverShield

DriverShield

Windows Kernel Driver Vulnerability & Malware Analysis Platform

Website API Status


Automated threat intelligence for Windows kernel drivers (.sys)

Detect vulnerabilities, malware patterns, rootkit indicators, and BYOVD (Bring Your Own Vulnerable Driver) attack vectors through deep static analysis, behavioral classification, and cross-referencing against known threat databases.


What is DriverShield?

image

DriverShield is a cloud-based analysis platform purpose-built for Windows kernel-mode drivers. Upload any .sys file and receive a comprehensive threat assessment covering code signing validation, dangerous API usage, IOCTL attack surfaces, rootkit behavior detection, known vulnerability matching, and behavioral risk scoring.

The platform is designed for security researchers, SOC analysts, incident responders, and threat intelligence teams who need fast, reliable driver-level analysis without setting up local toolchains.

Note: DriverShield is an automated analysis tool that provides risk indicators and heuristic assessments. Results should be evaluated by qualified professionals and do not constitute a definitive security verdict. See Disclaimer for details.


Analysis Capabilities

Core Engine — 14 Analysis Stages

Stage Description
Hash Identification MD5, SHA1, SHA256 fingerprinting with known-hash lookup
PE Header Analysis COFF headers, section entropy, version info, subsystem validation
Control Flow Integrity CFI/CFG, ASLR, DEP, integrity check detection with mitigations scoring
Code Signing Authenticode validation — signer identity, chain of trust, expiration
Import Analysis 80+ kernel APIs classified by risk level — flags dangerous functions commonly used in rootkits and privilege escalation
IOCTL Extraction Device I/O control codes with threat classification and known exploit flags
YARA Matching 12+ built-in behavioral signatures targeting rootkit patterns, process hiding, DKOM, SSDT hooking, and MBR/VBR manipulation
VirusTotal Automated hash lookup with per-engine detection results
LOLDrivers Cross-reference against the Living Off The Land Drivers database
CVE Mapping 20+ known vulnerable drivers mapped to CVE entries with CVSS scores
Behavioral Analysis Rule-based classifier detecting rootkit behavior, anti-AV evasion, persistence mechanisms, and privilege escalation patterns
Symbolic Execution angr-based path exploration with taint analysis (optional)
Sigma Rules Auto-generates 6+ SIEM detection rules per analyzed driver
MITRE ATT&CK Maps driver behavior to 16 ATT&CK techniques including Defense Evasion and Privilege Escalation

Rootkit & Stealth Detection

DriverShield applies heuristic analysis to identify common rootkit indicators in kernel drivers:

  • Direct Kernel Object Manipulation (DKOM) — Detects patterns associated with process/thread list unlinking
  • SSDT / IDT Hooking — Flags imports and code patterns related to system service descriptor table modification
  • I/O Request Packet (IRP) Hooking — Identifies file system and network filter driver patterns that may indicate interception
  • Registry Callback Abuse — Detects registration of notification callbacks commonly used for persistence and hiding
  • MBR/VBR Access Patterns — Flags low-level disk access associated with bootkits
  • Anti-Forensics — Identifies timestamp manipulation, log clearing, and trace removal techniques
  • Process & File Hiding — Detects behavioral patterns associated with hiding processes, files, or network connections from user-mode tools

These detections are heuristic-based and may produce false positives. Legitimate security software, anti-cheat systems, and system monitoring tools can trigger similar patterns. Always validate findings in context.

Risk Scoring

Dynamic weighted composite score (0-100) incorporating:

VirusTotal (22%)  |  API Risk (18%)  |  YARA (14%)  |  IOCTLs (13%)
CVE (10%)  |  LOLDrivers (9%)  |  Signatures (5%)  |  Dynamic (5%)  |  Packing (4%)

Verdicts: Clean (0-29) · Suspicious (30-59) · Vulnerable (60-79) · Malicious (80-100)


API

DriverShield provides a REST API for automated driver analysis workflows.

Hash Lookup

curl -s "https://drivershield.io/?r=apilookup&sha256=HASH"

Upload & Scan

curl -X POST "https://drivershield.io/?r=api/upload" \
  -H "X-Driver-MD5: YOUR_TOKEN" \
  -F "[email protected]"

Full API documentation: drivershield.io/api


Embeddable Badge

Show the analysis status of any driver on your website or README:

<img src="https://drivershield.io/?r=badge&sha256=DRIVER_SHA256" alt="DriverShield Badge">

Use Cases

Incident Response — Quickly assess whether a driver found during forensic investigation exhibits known threat characteristics or exploitable vulnerabilities.

SOC Triage — Integrate hash lookups into SIEM/SOAR workflows via the API to auto-enrich driver-related alerts with risk context.

Threat Intelligence — Track BYOVD campaign tooling, map driver artifacts to MITRE ATT&CK, and generate Sigma detection rules for proactive defense.

Vulnerability Research — Analyze IOCTL attack surfaces, dangerous kernel API usage, rootkit-associated patterns, and missing exploit mitigations.

Supply Chain Validation — Verify code signing integrity, check against LOLDrivers, and cross-reference CVE databases before deploying drivers in production environments.


Contact & Sponsorship

Support the Project

DriverShield is independently developed and maintained. If you find it useful, consider supporting its development:

Bitcoin: 19XiTXwVeh9hsY6dGbStCvYN6k3mAgEXHd


Disclaimer

DriverShield is provided "as is" for informational and research purposes only.

Analysis results are automated heuristic assessments and do not constitute a definitive security verdict, legal opinion, professional advice, or certification of any kind. The platform performs static analysis only — uploaded files are never executed, run, or detonated.

False positives and false negatives are expected. Legitimate software including anti-virus products, anti-cheat engines, hardware monitoring utilities, and system management tools routinely use privileged kernel operations that may trigger elevated risk scores. A high score does not mean a driver is malicious, and a low score does not guarantee a driver is safe.

Users are solely responsible for interpreting results and any actions taken based on them. The platform does not make accusations of wrongdoing against any software vendor, developer, or organization. Risk indicators reflect automated pattern matching against known threat characteristics and do not imply intent.

Users must ensure they have the legal right to upload and analyze any files submitted. DriverShield does not knowingly facilitate intellectual property theft, unauthorized reverse engineering, or any unlawful activity.

DriverShield, its developer, and any associated parties shall not be held liable for any direct, indirect, incidental, consequential, or special damages arising from the use of or inability to use this service, including but not limited to damages for loss of data, business interruption, or reputational harm.

This platform is not a substitute for professional security audits, penetration testing, code review, or legal compliance assessments. Organizations with regulatory requirements should engage qualified professionals for formal evaluation.

By using DriverShield, you acknowledge that you have read, understood, and agree to these terms.


Developed by Tolga SEZER

DriverShield — Vulnerability & Malware Analysis

About

Windows Kernel Driver Vulnerability & Malware Analysis Platform

Resources

License

Security policy

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors