#1Security papers, research projects, technical references, and reproducible material for defensive or authorized analysis.
Kitploit recommended

cve-2022-42889 Text4Shell CVE-2022-42889 affects Apache Commons Text versions 1.5 through 1.9. It has been patched as of Commons Text version 1.10.

Unified Security Research Tool

Proof-of-concept exploit and technical analysis for CVE-2026-48907, a CVSS 10.0 pre-authentication remote code execution vulnerability in the Joomla…

Detailed security advisory for CVE-2026-36669: unauthenticated arbitrary file upload in Feng Office, enabling stored XSS and session hijacking.…

CVE-2026-28992 IOHIDFamily FastPathUserClient race condition PoC — security research

Demonstrates CVE-2021-21300 arbitrary code execution via malicious Git hooks on case-insensitive filesystems, including exploit steps and defensive…

Research code for measuring jailbreak attack success rates (ASR and stepwise ASR) against LLMs, with implementations of attacks such as GCG and…

Session-scoped TOTP rate limiting permits repeated verification attempts

Local privilege escalation PoC for CVE-2026-74469, exploiting an SCTP diagnostic-reply kernel memory corruption to gain root on Linux x86-64 systems.

Claude Code skill for reverse-engineering 32-bit little-endian x86 C++ binaries (vtables, RTTI, inheritance recovery)

Defensive vulnerability-research project comparing vulnerable and patched Grandstream GXP1600 firmware for CVE-2026-2329, using SquashFS extraction,…

Interpretability and explainability of data and machine learning models

A deep learning toolkit for log-based anomaly detection

Explainable security gate for LLM apps — blocks prompt injection with an auditable reason for every decision.

Apache ActiveMQ Classic RCE research: CVE-2026-34197 / CVE-2026-42588 bypass chain + hardened-6.2.6 audit findings + Crowdfense comparison

Code for ACL 2026 (main) paper "DeepGuard: Secure Code Generation via Multi-Layer Semantic Aggregation"

Technical vulnerability analysis and CVE briefing for CVE-2026-9645 affecting ScadaBR.

Open standard for documenting security-relevant metadata of AI models, including training data provenance, PII risk, known vulnerabilities, and…