agentic-ai-security

A secure low code deception runtime framework, leveraging AI for System Virtualization.

An open specification for agentic AI security evaluation and testing, from Cisco.

AI Security Platform: Defense (61 Rust engines + Micro-Model Swarm) + Offense (39K+ payloads)

AI/ML and Generative AI Security Assessment Framework for AWS. Automatically audit Amazon Bedrock and SageMaker & AgentCore workloads for security best practices

Static AI Agent Risk Analyzer for AI applications. Detects capabilities, prompt risks, tool permissions, MCP integrations, governance gaps, and autonomy before deployment. Generate JSON, HTML and SARIF reports for CI/CD.

💰 Exocomp Agentic Environment for Go

Agentic AI Security Bootcamp is a hands-on, research-driven training environment for analysing, attacking, and securing autonomous AI systems. The repository provides structured labs, adversarial evaluation frameworks, and red-teaming exercises covering multi-agent observability, prompt injection..

Security working agreements for AI coding agents: hardened AGENTS.md, prompt/tool-injection guardrails, dependency hygiene, Scorecard-ready OSS setup

Claude code skills, agents, memory, profiles to accomplish cybersecurity tasks, projects, jobs with ease. claude-code, claude-plugin, claude-code-plugin, marketplace, security, genai-security.

MLSecOps Practical Reference Guide, open-source AI and ML security handbook.

🤖 Test and secure AI systems with advanced techniques for Large Language Models, including jailbreaks and automated vulnerability scanners.

The open standard for AI agent integrity. Evaluate, enforce, and prove that autonomous agents are adversarially coherent, environmentally portable, and verifiably assured.

Practitioner-led AI security control framework: 57 controls across 12 NIST AI 600-1 GenAI risk domains, mapped to MITRE ATLAS, in three tiers. Vendor-agnostic, CC BY 4.0.

TypeScript/JavaScript SDK for AI Agent Security - Drop-in security for LangChain, CrewAI, AutoGPT and custom agents

Essays on agentic AI security, decision-rights, reversibility-graded authority, manifest-declared action class, deterministic gates, and standards contribution method.

BioOS Cyber Genesis Challenge: An interactive web sandbox proving the 100% security paradigm of the Causal Operating System. Experience "Digital Causal Closure" firsthand: a world where hacking is a mathematical impossibility. Includes a vulnerable app protected by Z3 formal logic and hardware-validated intent (IRQ). Unhackable by design.

Risk-Aware Introspective RAG (RAI-RAG) is a safety-aligned RAG framework integrating introspective reasoning, risk-aware retrieval gating, and secure evidence filtering to build trustworthy, robust, and secure LLM and agentic AI systems.

Formal safety framework for AI agents. Pluggable LLM reasoning constrained by mathematically proven budget, invariant, and termination guarantee. 7 theorems enforced by construction, not by prompting. Includes Bayesian belief tracking, causal dependency graphs, sandboxed attestors, environment reconciliation, and a 155-test adversarial suite.

Open-source registry layer for auditable AI agent systems.

Website for the AVE open standard — the behavioral vulnerability classification standard for Agentic AI components.