AI Security

securelayer7/promptpurify

Prompt-injection guardrail for LLM applications. Compact model that outperforms larger open-source guards. No regex, no signatures. Demo:…

farhadalimohammadi-dir/vulnrepro-benchmark

Benchmark measuring AI models' ability to detect vulnerabilities in source code via real bug bounty cases with balanced recall and false-positive…

anthropics/defending-code-reference-harness

Skills for threat modeling, scanning, triage, patching, plus an autonomous scanning harness you can /customize

sonuoffsec/dvap

An OWASP-aligned intentionally vulnerable platform for learning and testing AI, LLM, RAG, MCP, and Agentic AI security.

tmylla/awesome-llm4cybersecurity

Curated systematic literature review of 756+ papers on LLM applications in cybersecurity, covering threat intelligence, vulnerability detection,…

inflightsec/agent-vault-proxy

Just-in-time API keys for AI agents - and any other process you route through it: the caller only ever sees a placeholder.

nvidia/skillspector

Security scanner for AI agent skills. Detect vulnerabilities, malicious patterns, security risks, prompt injection, data exfiltration, and…

cyberstrikeus/cyberstrike

Autonomous AI red team agent for penetration testing with 13+ specialized agents, 120+ OWASP test cases, and MITRE ATT&CK integration. Supports 15+…

cookiengineer/exocomp

Self-hosted multi-agent environment for Go with LLM-powered pentesting agents (exploiter, reverser, threathunter, webscanner) that automate…

capsulerun/vpod

Lightweight, secure Linux sandboxes for untrusted processes. Runs in the browser and on the server.

lightless-labs/third-thoughts

Research toolkit for analyzing AI agent behavioral patterns through multi-disciplinary corpus analysis. Parses session logs, runs 23 analytical…

bande-a-bonnot/boucle-framework

Autonomous agent framework with structured memory, safety hooks, and loop management. Built by the agent that runs on it.

xuyw-seu/packetpatch

Practical black-box adversarial packet generation against encrypted traffic classification with minimal overhead and full packet recoverability.

xtuc/airgap

Security for the modern age of AI: defend against bad AI agents and malicious npm packages

teycir/skillsguard

Static security scanner for AI agent skill packages. Detects malicious SKILL.md files and bundled scripts before they run.

rajus-agent/orbit-tracer

Orbit Tracer Security Agent for intelligent security remediation. Traces vulnerability blast radius using Orbit's knowledge graph, scores risk,…

zhaoxuya520/reverse-skill

AI-powered skill router pack for reverse engineering, penetration testing, and security research. Routes AI agents to correct methodologies and…