CVE-2024-37054
bardlaudian/cve-2024-37054Proof-of-concept that poisons MLflow registered models via the REST API, embedding a malicious pickle to trigger RCE when the model is loaded.
ai-securitydefensive-toolsexploitation+6
bardlaudian/cve-2024-37054Proof-of-concept that poisons MLflow registered models via the REST API, embedding a malicious pickle to trigger RCE when the model is loaded.
stardust457/sldrDefense framework that mitigates malicious fine-tuning of LLMs using selective layer recovery and dynamic routing, with training, harmfulness…
xtuc/airgapSecurity for the modern age of AI: defend against bad AI agents and malicious npm packages
shivang0/guardmodelGitHub Action that scans ML model files for malicious code and security vulnerabilities
db-ol/cuaharmBenchmark for evaluating safety risks of computer-using agents, with 104 realistic misuse scenarios across seven malicious categories, supporting…
andyshaw01/poisoncraftThis repository provides the official implementation of POISONCRAFT: Practical Poisoning of Retrieval-Augmented Generation for Large Language Models.
jpcertcc/logontracerInvestigate malicious Windows logon by visualizing and analyzing Windows event log
grassplatypus/cve-2026-45033-classCVE-2026-45033 PoC for Claude Code, not Github Copilot. Worked for Haiku 4.5.
graph-com/cka-agentResearch implementation of an LLM jailbreak agent that bypasses commercial model guardrails using harmless prompt weaving and adaptive tree search.
nvidia/skillspectorSecurity scanner for AI agent skills. Detect vulnerabilities, malicious patterns, security risks, prompt injection, data exfiltration, and…
hashgraph-online/hol-guardOpen-source antivirus for AI agents: block risky tools, secret access, prompt injection, malicious packages, MCP servers, plugins, and skills at…
promptfoo/modelauditSecurity scanner for AI/ML model files. Detects malicious code, backdoors, and vulnerabilities before deployment
kaill-jlq/mmskillriskBenchmark and evaluation harness testing whether LLM agents resist malicious instructions hidden in multimodal skill images, with 108 cases across…
gautamvarmadatla/mcpsafetywardenProxy server that wraps MCP servers with behavioral profiling, security scanning, risk gating, and safe execution. Detects prompt injection,…
teycir/skillsguardStatic security scanner for AI agent skill packages. Detects malicious SKILL.md files and bundled scripts before they run.
zhengxr930/malevalOfficial code for the ISSTA 2026 paper: Is "Knowing It’s Malicious" Enough? Evaluating LLMs for Fine-Grained Malware Behavior Auditing
yym8538/cve-2026-33017Proof-of-concept exploit for CVE-2026-33017, an unauthenticated RCE in Langflow's build_public_tmp endpoint, injecting a malicious custom component…
cisco-ai-defense/skill-scannerSecurity Scanner for Agent Skills