LeakGauge
yeasen-z/leakgaugeDetects LLM context-leakage attacks by training lightweight behavior probes on log-probabilities, with vLLM offline/server detection pipelines.
adversarial-attackai-securityanomaly-detection+2
yeasen-z/leakgaugeDetects LLM context-leakage attacks by training lightweight behavior probes on log-probabilities, with vLLM offline/server detection pipelines.
mthamil107/db_audit_researchSelf-Defeating Audits: reproducible lab showing a low-privilege PostgreSQL role reversibly blinding a trigger-based auditor + poisoning attribution…
cenobyte-vincit/xspawnSpawns macOS programs through launchd's private XPC interface without execing them, making EDR record launchd as parent. Supports one-shot,…
imabee101/cve-2023-44487PoC script for HTTP/2 Rapid Reset (CVE-2023-44487) that sends crafted HTTP/2 streams to trigger denial-of-service conditions on vulnerable servers,…
camillegr/cve-2026-73292Proof-of-concept exploit for CVE-2026-73292: CSRF attack on Semaphore UI password change endpoint, serving a malicious page that silently resets an…
lxxexxbxx/cve-2026-44578Proof-of-concept exploit for CVE-2026-44578 that reproduces the vulnerable condition, enabling security researchers to validate affected systems and…
gmh5225/awesome-ai-securityA curated list of AI Security materials and resources for Pentesters, Bug Hunters, and Security Researchers.
loosehose/dutchovenApplication-scoped Windows network brownouts in native C and BOF form
wattocyber/gradient-untanglerLocal white-box gradient attacks for open-weight LLMs: GCG/PEZ suffix search, layer saliency, weight snapshots, and rank-1 suffix-to-delta fitting…
defineid/skeletonkeyCVE-2026-6765 · Test only FormAutofill handlers exposed in Firefox
msnightmare/shieldbreakWindows Defender patch bypass PoC for CVE-2026-50656 (RoguePlanet), demonstrating exploitability on Windows 11 25H2 and Server 2025 despite…
aloshdenny/claude-awmEvades LLM text watermarks by injecting Unicode variation selectors; includes the SynthID generator, mean-g detector, normalization defenses, and…
rakib-nyc/nulloriginResearch-only AI watermark robustness toolkit: local reverse proxy strips C2PA/EXIF/XMP, Unicode, image/audio stego, OOXML/PDF metadata, and scans…
gmh5225/antidebug-antivmDifferent methods to detect a virtualized environment or potential debugging
xrazymee/ragsieveSelf-referenced local contrast for knowledge-poison detection in retrieval-augmented generation
guillaumemeyer/watermarks-removerStrip multi-vendor AI provenance marks: Unicode text hygiene, statistical rewrite hooks, and C2PA/metadata from PNG/JPEG/SVG/PDF/DOCX/HTML/MD
hunt-benito/the-sanitizer-is-the-weapon-cve-2026-68749-cve-2026-68750-quadratic-dos-in-elixir-html-sanitize-exPoC demonstrating quadratic DoS in Elixir html_sanitize_ex via crafted HTML; includes timing benchmarks, remote exploitation curl, and verification…
horkimhab/llm-prompt-injection-resourcesA curated collection of resources for learning and researching LLM prompt injection attacks, defenses, and security.