ContextHound

ContextHound Herramienta de análisis estático que escanea tu código base en busca de vulnerabilidades de inyección de prompts y seguridad multimodal. Funciona sin conexión, no requiere llamadas a API. El ecosistema de ContextHound ContextHound está disponible en todo tu flujo de trabajo de...

AutoRAN-public

🧠 AutoRAN: Secuestro automatizado del razonamiento de seguridad en grandes modelos de razonamiento AutoRAN es un secuestro automatizado del razonamiento de seguridad que aprovecha modelos auxiliares secundarios menos alineados para simular trazas de razonamiento, generar prompts narrativos y...

mythic_ornn

Mythic Ornn Generador impulsado por LLM para Agentes, Payload-Types y Perfiles C2 de Mythic. Los agentes de Mythic son tediosos de escribir a mano: cada uno reimplementa el mismo protocolo de red, el andamiaje de payload-type y la infraestructura de C2. Mythic Ornn omite el código repetitivo...

CVE-2026-78906-ChatGPT-Prompt-Injection

CVE-2026-78906 — PromptGhost ChatGPT API — Prompt Injection & Memory Exfiltration Field| Value ---|--- Severity| 8.7 High Vector| Network Affected Versions| API v1.0 – v1.3 Discovered By| 𝕍𝕠𝕤𝕤🥷 Description A critical vulnerability in OpenAI's ChatGPT API allows an attacker to inject malicious...

AgentWatcher

AgentWatcher AgentWatcher es una defensa basada en detección contra la inyección indirecta de prompts en agentes LLM. Primero ejecuta atribución causal de contexto sobre el contexto no confiable para encontrar los contextos más influyentes, y luego aplica un LLM monitor que clasifica esos context...

claude-project-scanner

Claude Project Scanner Read-only Claude Code skill that scans third-party projects for known security risks before you open them. Catches the attack patterns documented in: CVE-2025-59536 Claude Code Lifecycle Hooks injection CVE-2025-61260 OpenAI Codex CODEXHOME injection CVE-2025-54136 Cursor...

CVE-2026-100867

spaceship-prompt through 4.22.5 fails to sanitize control characters from project manifest version fields before rendering them in the zsh prompt. Attackers can embed ANSI/OSC escape sequences in version fields of package manifests to manipulate terminal output, rewrite window titles, or spoof...

EUVD-2026-87939

spaceship-prompt through 4.22.5 fails to sanitize control characters from project manifest version fields before rendering them in the zsh prompt. Attackers can embed ANSI/OSC escape sequences in version fields of package manifests to manipulate terminal output, rewrite window titles, or spoof...

CVE-2026-100867

Versions of spaceship-prompt up to and including 4.22.5 are vulnerable to Terminal Escape Sequence Injection . The software fails to sanitize control characters within project manifest version fields before rendering them in the zsh prompt . An attacker can embed ANSI/OSC escape sequences in thes...

CVE-2026-100867 spaceship-prompt through 4.22.5 Terminal Escape Sequence Injection

spaceship-prompt through 4.22.5 fails to sanitize control characters from project manifest version fields before rendering them in the zsh prompt. Attackers can embed ANSI/OSC escape sequences in version fields of package manifests to manipulate terminal output, rewrite window titles, or spoof...

CVE-2026-1337-AI-Coding-Assistant-Prompt-Injection-to-Sandbox-Escape

CVE-2026-1337 – AI Coding Assistant Prompt Injection to RCE 📖 Overview A prompt injection vulnerability in an AI-powered code review bot allows an attacker to inject arbitrary shell commands by manipulating the bot’s “fix” suggestion. The sanitizer uses a simple blacklist that is bypassed using...

rebuff

Rebuff.ai Detector de inyección de prompts con auto-endurecimiento Rebuff está diseñado para proteger aplicaciones de IA contra ataques de inyección de prompts PI mediante una defensa multicapa. Playground • Discord • Características • Instalación • Primeros pasos • Autohospedaje • Contribuciones...

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...

cve-2026-54316-lab

CVE-2026-54316 — Laboratorio de exfiltración de Claude Code vía WebFetch y HuggingFace Un laboratorio autocontenido y desechable que reproduce GHSA-fg94-h982-f3mm / CVE-2026-54316: Claude Code preaprobó huggingface.co como un nombre de host simple para la herramienta WebFetch, por lo que cualquie...

llm-prompt-injection-resources

llm-prompt-injection-resources Una colección seleccionada de recursos para aprender e investigar sobre ataques de inyección de prompts en LLM, defensas y seguridad. Donar Apoya el mantenimiento de este proyecto con PayPal o escaneando el código QR a continuación...

When Consent Outlives Context: Residual Authority Replay in Long-Lived Agents

LLM agents increasingly rely on user approval to authorize security-sensitive actions at runtime. Such approvals are granted within a specific task and execution context. In long-lived agents, authorization decisions may need to persist across tasks or sessions. We find that this continuity can...

Can Prompt Anonymity Protect Your Identity from LLM Providers?

User conversations with large language models LLMs often contain highly sensitive personal information that can be exploited by LLM providers to create detailed user dossiers, enable targeted advertising, and train more powerful models. To protect user privacy, anonymizing LLM proxies have emerge...

Prompt-Injection-in-the-Wild

Prompt Injection in the Wild A tracker of publicly reported prompt-injection techniques from roughly the last two years, maintained by Rachel James cybershujin. Each technique is broken down into the three elements of a prompt injection: 1. Delivery method — how the injected instructions reach th...

CVE-2026-100651

A flaw was found in vLLM. The service fails to properly enforce length limits on input prompts when handling certain multimodal inference requests. A remote attacker with access to the inference service could exploit this vulnerability by submitting an excessively long prompt, causing an internal...

EUVD-2026-87745

vLLM before 0.29.0 fails to enforce decoder prompt-length validation on the disaggregated serving endpoint /inference/v1/generate. When the request contains a 'features' multimodal payload, vllm/entrypoints/serve/disagg/serving.py builds a multimodal EngineInput directly from the caller-supplied...