wardgate

Wardgate - AI Agent Security Gateway Wardgate is a security gateway that sits between AI agents and the outside world -- isolating credentials for API calls, isolating SSH keys for remote command execution, and gating command execution in remote environments conclaves. Give your AI agents access ...

node9-proxy

🛡️ Node9 What did your AI agent actually do? Find out. Node9 sits between your AI agent and the tools it can use — what it's already been doing, against risky actions in real time, and what happened over any time window...

sigma-ai

AgentShield Sigma Rules What is This Repository? This repository contains detection rules that help identify when an AI agent is being attacked or manipulated. Think of it as a library of "threat signatures" -- each rule describes a pattern that, when matched against an agent's log data, signals...

tealtiger

TealTiger AI Agent Security & Governance SDK Deterministic governance, guardrails, cost tracking, and policy management for LLM applications. Open source. TypeScript + Python. Works with any provider...

MakerChecker

🛡️ MakerChecker The open-source security layer for AI agents. Deny-by-default enforcement, human approvals, and a cryptographically signed audit trail — so your agent runs only what it's granted and provably can't approve its own work. Website · Live Demo · mc scan · Docs Your agents keep running...

sage

Sage Safety for Agents — Agent Detection & Response for AI coding assistants Sage is a lightweight security layer that protects AI agents from executing dangerous actions. It intercepts tool calls — shell commands, URL fetches, file writes — and checks them against multiple threat detection layer...

hush

Portable, open specification for AI agent security rules Spec · Docs · Rulesets · JSON Schema HushSpec is an open policy format for AI agent security rules. It defines what an agent may do at runtime, including filesystem access, network egress, tool usage, secret detection, and more, without...

secure-agent-playbook

Secure Agent Playbook AI 에이전트를 위한 오픈소스 보안 플레이북입니다. 구조화된 OWASP 기반 절차를 통해 에이전트가 코드 리뷰부터 AI 에이전트 보안 감사까지 보안 엔지니어링 작업을 수행할 수 있게 합니다. 목차 왜 사용하나요? 이것은 무엇인가? 빠른 시작 스킬 카탈로그 에이전트 예시 출력 플레이 아키텍처 OWASP 재단 관련 프로젝트 기여 왜 사용하나요? 플레이북 없이 AI 에이전트에게 "내 코드를 보안 관점에서 검토해줘"라고 요청하면 피상적인 체크리스트만 얻을 수 있습니다. 그러나 이 플레이를 사용하...

Agent-Security-Regression-Harness

OWASP Agent Security Regression Harness OWASP Agent Security Regression Harness는 에이전트 애플리케이션 및 MCP 통합 시스템에 대해 실행 가능한 보안 회귀 시나리오를 실행하기 위한 오픈 소스, 벤더 중립적 테스트 하네스입니다. 이 프로젝트는 빌더와 방어자가 프롬프트, 모델, 도구, 검색 소스, 메모리, 승인 흐름 또는 MCP 통합에 대한 변경으로 인해 알려진 보안 실패가 재발하지 않는지 검증할 수 있도록 도와줍니다. 이 프로젝트가 하는 일 이 프로젝트는 다음과 같...

aegisagent

AegisAgent — AI Agent Runtime Security Gateway AI Agent 런타임 보안 게이트웨이 — 모든 도구 호출을 실제 모델 생성 결과에 암호학적으로 바인딩하고, 제로 트러스트 정책 시행, 샌드박스 격리와 인간 승인HITL을 통합합니다. 배경: 왜 AegisAgent가 필요한가 2026년 8월, Cloud Security Alliance는 CoreBreak 취약점군 을 공개했습니다. 이 일련의 심각한 취약점들은 AI Agent 생태계의 보안 가정을 완전히 깨뜨렸습니다: CVE 번호| 영향 컴포넌트...

clawguard

🦞 ClawGuard v3 엔터프라이즈 AI 에이전트 보안 툴킷 - SKILL.md 기반 능동 방어 핵심 개념 ClawGuard v3의 핵심 방어는 코드가 아니라 SKILL.md에 있습니다! 각 모듈의 SKILL.md 자체가 완전한 방어 가이드입니다: 에이전트에게 언제 트리거할지 알려줍니다 에이전트가 어떻게 탐지할지 안내합니다 구체적인 탐지 패턴과 규칙 을 제공합니다 출력 형식과 판단 기준 을 정의합니다 코드 스크립트는 보조 도구일 뿐이며, 실제 지능은 SKILL.md에 있습니다. 🎯 5가지 보안 모듈 모듈| 위치|...

CVE-2026-40145

A vulnerability exists in the interaction between a Endpoint Privilege Management Windows Deployment support utility and the agent's tamper protection controls. Under certain conditions, the protections applied to the utility process may not be enforced as intended...

CVE-2026-57282

Jenkins Git client Plugin 6.6.0 and earlier does not correctly escape the workspace directory name when it is embedded into a generated SSH wrapper script, allowing attackers able to control the name of a build's working directory to execute arbitrary operating system commands on the agent...

-cascade-scan

cascade-scan AI Agent security evaluation framework — autom...

AI Agent Security: Automating Workflow Without Creating Prompt Injection or Data Leak Risks

AI agent security starts with a simple fact: the more authority an agent has, the tighter its access…...

Agent Security Is a Systems Problem

We take the position that agent security must be approached as a systems problem: the AI model powering the agent must be treated as an untrusted component, and security invariants must be enforced at the system level. Through this lens, efforts to increase model robustness the dominant viewpoint...

Architecture Matters for Multi-Agent Security

Multi-agent systems MAS, composed of networks of two or more autonomous AI agents, have become increasingly popular in production deployments, yet introduce security risks that do not arise in single-agent settings. Even if individual agents exhibit robust security, architectural decisions...

Ghost in the Agent: Redefining Information Flow Tracking for LLM Agents

Autonomous Large Language Model LLM agents are increasingly deployed to conduct complex tasks by interacting with external tools, APIs, and memory stores. However, processing untrusted external data exposes these agents to severe security threats, such as indirect prompt injection and unauthorize...

Towards Modeling Cybersecurity Behavior of Humans in Organizations

We undertake a comprehensive and structured synthesis of the drivers of human behavior in cybersecurity, focusing specifically on people within organizations i.e., especially employees in companies, and integrate key concepts such as awareness, security culture, and usability into a coherent...

Targeted Bit-Flip Attacks on LLM-Based Agents

Targeted bit-flip attacks BFAs exploit hardware faults to manipulate model parameters, posing a significant security threat. While prior work targets single-step inference models e.g., image classifiers, LLM-based agents with multi-stage pipelines and external tools present new attack surfaces,...