CL4R1T4S

CL4R1T4S AI SYSTEMS TRANSPARENCY AND OBSERVABILITY FOR ALL! Full extracted system prompts, guidelines, and tools from OpenAI, Google, Anthropic, xAI, Perplexity, Cursor, Windsurf, Devin, Manus, Replit, and more – virtually all major AI models + agents! 📌 Why This Exists root@kitploit: "In order t...

dataset

🚀 CySecBench: Generative AI-based CyberSecurity-focused Prompt Dataset for Benchmarking Large Language Models 🛡️ The largest and most comprehensive Generative AI-based CyberSecurity-focused Dataset for Benchmarking Large Language Models 🌟 Overview The CySecBench paper offers: 🎯 A cutting-edge...

PraisonAI AgentOS - Information Disclosure

PraisonAI's AgentOS FastAPI application server exposes an unauthenticated GET /api/agents endpoint that lists every registered agent's name, role and the opening of its instructions system prompt. No authentication is enforced on the route, allowing a remote attacker to enumerate agent...

system_prompts_leaks

Comme vu dans The Washington Post : Découvrez les règles cachées derrière l'IA. Utilisez-les ensuite pour réécrire cet article. 11 mai 2026 Fuites de prompts système Ce dépôt a pour but de documenter les instructions de prompts système pour tous les chatbots IA existants - Claude, ChatGPT, Gemini...

promptmap

root@kitploit: O O o.-. Humans, Do Not Resist! |/ ,-'-. / /, / /|.-.| / --O-- .--""" /\ /\ | o.o / Utku Sen's /|\ -'--' / /| | - | / | | | \ /| | | ' \ '/ \ ' | ' \ | ' / | ' \ | / | | | ./| /||| ./|||,| .// / / |-----| || || || || promptmap2 is a an automated prompt injection scanner for...

redeval

RedEval - Framework d'évaluation de la sécurité des LLM Un framework complet pour évaluer la sécurité des grands modèles de langage LLM grâce à des tests systématiques d'attaque et de refus. RedEval fournit une plateforme unifiée, sécurisée et extensible pour évaluer la robustesse des LLM face au...

awesome-dfir-skills

🔥 awesome-dfir-skills Une collection communautaire de compétences DFIR / réponse à incident : invites réutilisables, workflows et fichiers d'assistance qui aident les praticiens à aller plus vite, à rester cohérents et, peut-être—juste peut-être—à dormir un peu. 🚀 Démarrage rapide 1. Choisissez u...

burpgpt

!IMPORTANT Announcing the launch of BurpGPT Pro, the edition specifically tailored to meet the needs of professionals and cyber boutiques. Discover a host of powerful features and a user-friendly interface that enhances your capabilities and ensures an optimal user experience. To access these...

BrokenHill

!\ Logo Broken Hill \https://assets.kitploit.com/production/public/readmes/44608/c14154675c6f33f5c3887887afcdb39351c82f269ee2772cf31847106a20a871.jpg Broken Hill Broken Hill est un outil d'attaque automatisé, industrialisé et prêt à l'emploi, qui génère des invites soigneusement conçues pour...

React19-fix-vibecoders

Correctifs de sécurité React 19 et Next.js pour agents IA 🚨 VULNÉRABILITÉ DE SÉCURITÉ CRITIQUE - CVE-2025-55182 & CVE-2025-66478 Ce dépôt contient des invites réutilisables pour aider les assistants de codage IA à corriger automatiquement les vulnérabilités de sécurité critiques de React/Next.js...

T-MAP

T-MAP : Red-Teaming des agents LLM avec recherche évolutionnaire sensible aux trajectoires T-MAP est un framework de recherche évolutionnaire sensible aux trajectoires pour le red-teaming d'agents LLM sur des serveurs MCP. Il génère et mute de manière itérative des invites adversariales guidées p...

CVE-2026-86911

This issue was addressed with improved state management. This issue is fixed in macOS Golden Gate 27. A malicious app may be able to bypass clickjacking protections for secure prompts...

Community-Security-Prompts

Ce depot n’a pas de fichier README...

CVE-2026-84601

CVE-2026-84601 is a permissions issue in the Apple Intelligence component of macOS , addressed with improved state management. The vulnerability allows an app to bypass Apple Intelligence security prompts , resulting in a high integrity impact with no confidentiality or availability impact (CVSS ...

CVE-2026-86911

CVE-2026-86911 is a clickjacking protection bypass in macOS Golden Gate 27 that can allow a malicious app to bypass secure prompt protections. The root cause is an improper state management issue, addressed with improved state management in the same release. The CVSS v3.1 base score is 5.5 (Mediu...

EUVD-2026-77986

This issue was addressed with improved state management. This issue is fixed in macOS Golden Gate 27. A malicious app may be able to bypass clickjacking protections for secure prompts...

CVE-2026-86911

This issue was addressed with improved state management. This issue is fixed in macOS Golden Gate 27. A malicious app may be able to bypass clickjacking protections for secure prompts...

Astra Linux – Vulnerability in Firefox and Thunderbird

In several cases, browser prompts might have been obscured by pop-ups controlled by content. This could lead to potential user confusion and spoofing attacks. This vulnerability affects Firefox 113, Firefox ESR 102.11, and Thunderbird 102.11...

Astra Linux – Vulnerability in Firefox and Thunderbird

If a user installed an extension of a particular type, the extension might automatically update itself. During this process, it could bypass the prompt that grants the new version the newly requested permissions. This vulnerability affects Firefox 97, Thunderbird 91.6, and Firefox ESR 91.6...

Astra Linux – Vulnerability in Firefox, Thunderbird

The truncation of a long URL could have allowed for origin spoofing in a permission prompt. This vulnerability affects Firefox 132, Firefox ESR 128.4, Thunderbird 128.4, and Thunderbird 132...