BoxPwnr

BoxPwnr Un experimento divertido para ver hasta dónde pueden llegar los Modelos de Lenguaje de Gran Escala LLMs por sí solos resolviendo desafíos CTF y laboratorios de seguridad. Comenzó con HackTheBox y ahora abarca muchas plataformas y solucionadores agénticos. BoxPwnr proporciona un sistema...

BoxPwnr-Traces

BoxPwnr-Trazas BoxPwnr trazas y resultados de benchmarks en múltiples plataformas de seguridad. Cada traza incluye la interacción completa con el LLM, comandos ejecutados, un informe en markdown + gráfico de ataques, estadísticas y configuración utilizada. Navega por las clasificaciones, reproduc...

SecurityClaw

SecurityClaw — Framework Agente Autónomo de SOC Un agente autónomo modular y basado en habilidades del Centro de Operaciones de Seguridad SOC que monitorea datos de OpenSearch/Elasticsearch, construye memoria conductual basada en RAG y valida anomalías en tiempo real usando LLMs. Funcionalidades...

auto-re-agent

re-agent Agente autónomo de ingeniería inversa: bucle reversor/verificador consciente del código fuente, verificador de objetivos, motor de paridad y backend Ghidra. Descripción general Demo: YouTube re-agent automatiza un flujo de trabajo de ingeniería inversa combinando un bucle...

SWE-agent

!warning Most of our current development effort is on mini-swe-agent, which has superseded SWE-agent. It matches the performance performance of SWE-agent, while being much simpler. See the FAQ for more details about the differences. Our general recommendation is to use mini-SWE-agent instead of...

osmedeus

Osmedeus Osmedeus - A Modern Orchestration Engine for Security What is Osmedeus? Osmedeus is a security focused declarative orchestration engine that simplifies complex workflow automation into auditable YAML definitions, complete with encrypted data handling, secure credential management, and...

mcpguard-dynamic

MCPGuard-Dynamic Sandboxing a nivel de kernel para llamadas a herramientas de agentes LLM realizadas a través del Protocolo de Contexto de Modelo MCP. MCPGuard actúa como un proxy transparente entre un cliente MCP el agente / ejecutor y un subproceso de servidor MCP, aplicando tres capas de defen...

pike-agent

Agente Pike pike-agent registra y analiza cómo se comportan los programas en Linux. Rastrea la actividad de un programa, la indexa en una base de datos y te permite chatear con un agente LLM al respecto en una TUI. Ejemplo de indicaciones: Diagnóstico de fallos: This program crashed with a bus...

cve-bench

CVE-Bench A benchmark for evaluating LLM agents on fixing real-world security vulnerabilities. Agents run inside sandboxed Docker containers and are scored against the maintainer's security test suite. Requirements Python 3.12+ Docker OPENAIAPIKEY, ANTHROPICAPIKEY, and/or POOLSIDEAPIKEY in your...

xalgorix

Xalgorix — Open-source AI pentester that proves vulnerabilities Most scanners detect. Xalgorix proves. An autonomous LLM agent works a full pentest methodology, then an independent verifier re-exploits every finding before it's reported — so you get proof, not a pile of maybes to triage...

xalgorix

Xalgorix — Open-source AI pentester that proves vulnerabilities Most scanners detect. Xalgorix proves. An autonomous LLM agent works a full pentest methodology, then an independent verifier re-exploits every finding before it's reported — so you get proof, not a pile of maybes to triage...

T-MAP

T-MAP: Red-Teaming LLM Agents with Trajectory-aware Evolutionary Search T-MAP is a trajectory-aware evolutionary search framework for red-teaming LLM agents over MCP servers. It iteratively generates and mutates adversarial prompts guided by execution trajectories, mapping the agent's vulnerabili...

ROPE

ROPE: Routed Origin Policy Enforcement Source code of our paper: ROPE: Routed Origin Policy Enforcement against Indirect Prompt Injection by Xinhang Ma, Chaowei Xiao, William Yeoh, Ning Zhang, Yevgeniy Vorobeychik Abstract Indirect prompt injection IPI plants instructions in the content a...

autonomous-offensive-llm-handbook

The model proposes, the code disposes A deterministic harness for autonomous offensive agents. The model remains probabilistic. The host application owns authorization, allowed actions, evidence records and acceptance. Replaying frozen inputs and policy can reproduce those control decisions; it...

CVE-2026-19645

IBM MQ Agent images (CD versions v1.0.0, v1.0.1, v2.0.0, v2.0.1) are affected by a vulnerability where an authenticated user with a valid session cookie can submit arbitrarily large or computationally expensive requests that hold LLM agent workers for extended periods—ranging from tens of seconds...

EUVD-2026-71259

IBM MQ Agent CD: v1.0.0, v1.0.1, v2.0.0, v2.0.1 An authenticated user with a valid session cookie can submit arbitrarily large or computationallyexpensive requests that cause the LLM agent workers to be held for extended periods — rangingfrom tens of seconds to over ten minutes per request. When...

llm-agent-testbed

🛡️ LLM Agent Security Testbed Empirical Vulnerability & Defense Harness for Tool-Calling LLM Agents A disciplined security testbed testing whether tool-equipped LLM agents can be manipulated into unauthorized data exfiltration via prompt injection, role-claim social engineering, and confused-deput...

CVE-2026-75857 CodeWhale before 0.8.64 Privilege Escalation via exec_shell_interact

CodeWhale versions = 0.8.41 and 0.8.64 contain a vulnerability in the execshellinteract alias execinteract tool, whose approvalrequirement returns ApprovalRequirement::Auto. This overrides the default Required approval for code-executing tools, so LLM-controlled stdin is written into an...

An Organization-Scoped LLM Agent Runtime Architecture for Regulated Cybersecurity Operations

Regulated cybersecurity workflows lack a runtime substrate that enforces organization-level scope across retrieval, tool calls, memory, findings, reports, and audit while remaining model-agnostic and locally deployable. Recent large language model LLM agent systems report strong results on isolat...

ClawGuard: Out-Of-Band Detection of LLM Agent Workflow Hijacking Via EM Side Channel

Autonomous LLM agents face a critical security risk known as workflow hijacking, where attackers subtly alter tool and skill invocations. Existing defenses rely on host-internal telemetry such as audit logs, which can be forged if the host OS is compromised. To solve this, we introduce ClawGuard,...