authproof-sdk

AuthProof SDK AuthProof 是一种面向代理型 AI 的加密授权协议。该领域的大多数协议都基于操作员定义的策略执行——赋予操作员在事后扩展或重新解释用户原始意图的权力。AuthProof 基于不同的信任模型构建：用户自己的私钥签署控制执行的授权对象，并且在授权时和执行前立即验证实时模型状态。用户签署的权限与实时模型状态门控的结合是具体的主张——而非广泛的执行叙事。 不同之处在于： 用户是签署权威。 每个竞争协议（AIP、AITH、OAP、SAGA、AgentSpec）都基于操作员定义的策略执行。在 AuthProof...

llm-agent-testbed

🛡️ Banco de Pruebas de Seguridad para Agentes LLM Banco de Pruebas Empírico de Vulnerabilidades y Defensas para Agentes LLM con Llamada a Herramientas...

nexus-os

Nexus OS 受控的自主智能体操作系统 66 个 crate | 675 个命令 | 86 个页面 | 5229 个测试 | OWASP 10/10 | 零桩 本地优先。可断网隔离。后量子就绪。用 Rust 构建。 架构 | 快速开始 | 特性 | 审计状态 | 文档 Nexus OS 是一个 AI 智能体操作系统，其中智能体拥有一等公民地位，具备密码学身份、受控自主权以及演进能力。它完全运行在你的硬件上——无云端依赖，数据不离开你的机器，可断网隔离。每个操作都通过哈希链记录，每个决策都可审计，每个智能体都运行在沙箱中。 root@kitploit:...

mcp-attack-detection-sentinel

MCP Attack Detection — Sentinel Research Lab Sentinel detection research for MCP Model Context Protocol attack patterns : unusual identity access after an MCP-related event, tool-definition mutation, cross-resource access, and possible post-exploitation behavior. The rules map to the OWASP Top 10...

mulot

mulot -4285F4?logo=googlechrome&logoColor=white Agentic AI web pentester that drives a browser. An open-weights LLM GLM-5.2, Gemma or Qwen drives a real headless Chromium through a Burp-style toolkit and works a target the way a human pentester would. No frontier model, no agent running inside a...

agentic-ioc-scanner

agentic-ioc-scanner 针对自主AI编码工具的IOC扫描器——检测 Mini Shai-Hulud、Gemini CLI RCE、Cursor CVE-2026-26268 以及 DPRK PromptMink。 一个检测工具包，用于检测自主AI编码助手（Claude Code、Gemini CLI、Cursor）及其拉入仓库的依赖项是否受到危害。共十一个检查项，涵盖钩子注入、RCE配置、恶意依赖、Git钩子后门以及CI工作流篡改。IOC列表外部化——添加新指标无需修改代码。 配套博客文章：When the Tool Fights Back。 包含内容 1. IOC扫描器...

isa_recovery

Sistema de Recuperación ISA Un pipeline de ingeniería inversa que convierte un binario de firmware y su desensamblado posiblemente incorrecto en una especificación de procesador funcional para Ghidra. Cuando te encuentras con un procesador propietario sin documentación y sin soporte en Ghidra, es...

Agentic-CLIP-Benchmark

Agentic-CLIP-Benchmark: 在CIFAR-10上的零样本评估 -green.svg 本项目实现了一个稳健的自动化管道，用于在完整的 CIFAR-10 测试集（10,000张图片）上评估OpenAI的 CLIP ViT-B/32 模型。采用 AI原生工作流（Trae IDE） 开发，实现了高精度的零样本准确率 88.80% 。 📊 性能总结 Top-1准确率 : 88.80% 零样本 模型 : openai/clip-vit-base-patch32 使用Safetensors 推理硬件 : NVIDIA RTX 2060 时间复杂度 : 通过批推理优化（批大小：32）...

IPI-exposure-signal

IPI Exposure Signal This is the code repository for our paper: Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure. ArXiv version and paper link: https://arxiv.org/abs/2608.02657 This repository implements the probing pipeline for latent IPI-exposure signals:...

KidnapRAG

KidnapRAG: Un Ataque de Caja Negra para Secuestrar el Razonamiento en Sistemas de Generación Aumentada por Recuperación RAG Basados en Agentes 🎓 Paper | 📄 Datasets | 🚀 Inicio Rápido Inicio Rápido Ejecute los siguientes comandos desde el directorio padre del repositorio clonado KidnapRAG...

CVE-2025-59536

CVE-2025-59536 - la implementación del diálogo de confianza de inicio. Vulnerabilidad de Claude Gravedad: ALTA CVSS: 8.8 Impacto: Confidencialidad, Integridad, Disponibilidad Publicado: 2025-10-03 Legal Solo para pruebas de seguridad autorizadas. Causa raíz versión corta Claude Code es una...

agentic-dm-gateway

Agentic DM Gateway Security control plane for LLM agents over private chat typically Discord DMs. It sits in front of your agent. It decides who may talk, whether the session is unlocked, whether the process is paused, and whether this message is safe enough to forward. Your model and tools stay...

violin

Violin ☤ — Perfil de Pentest Agéntico Supervisado para Hermes 35 playbooks · 17 referencias · 13 plantillas · guarda de ejecución obligatoria · nativo de Hermes Violin es un perfil de pentest agéntico nativo de Hermes para pruebas de penetración supervisadas y autorizadas — desde el reconocimient...

vigolium v0.4.4

Vigolium - High-fidelity vulnerability scanner fusing agentic AI with native speed, modularity, and precision Vigolium provides two complementary scanning modes: Native Scan vigolium scan: Fast, powerful, and flexible. Deterministic, multi-phase scanning with 317 modules across content discovery,...

violin v3.2.1

Violin ☤ — Supervised Agentic Hermes Pentest Profile 35 playbooks · 17 references · 13 templates · required execution guard · Hermes-native Violin is a Hermes-native agentic pentest profile for supervised, authorised penetration tests — from reconnaissance through safe exploit validation to...

Amazon Kiro Prompt Injection Can Exfiltrate Sensitive Data Through Kiro Powers

Cybersecurity researchers have disclosed details of a vulnerability in Amazon Kiro, an artificial intelligence AI-powered, agentic integrated development environment IDE, that could facilitate data exfiltration via prompt injection and Kiro Powers. The security flaw, which does not have a CVE...

The OWASP LLM Top 10 Was the Warm-Up: What Comes Next

This is Post 3 of a three-part series. Start withPost 1: Agentic AI Security: The Chatbot Era Is Over, then continue to Post 2: Generative AI Security: Why AI Needs a New Kind of Security. When the OWASP Top 10 for LLM Applications arrived, it did the industry a real service. It gave security tea...

bromure agentic-coding-v4.5.4

Bromure Secure, ephemeral computing in disposable Linux VMs on macOS. → Full details, screenshots, and downloads at bromure.io This repo ships two sibling apps, both built on Apple's Virtualization.framework: Bromure — every browser session runs in a throwaway Linux VM. Close the window, the VM i...

CVE-2026-9196

IBM Langflow OSS 1.0.0 through 1.10.3 could allow an authenticated attacker to execute unintended code during Agentic Assistant validation due to improper handling of LLM‑generated components. The application executes model‑generated Python code in the backend during validation prior to user...

CVE-2026-9196

IBM Langflow OSS 1.0.0 through 1.10.3 could allow an authenticated attacker to execute unintended code during Agentic Assistant validation due to improper handling of LLM‑generated components. The application executes model‑generated Python code in the backend during validation prior to user...