owasp-llm-top10

🛡 The reference playbook for pentesting AI chatbots & LLM-powered apps in one place. Ready-to-use payloads covering the full OWASP LLM Top 10 plus frontier vectors (MCP · RAG · A2A · computer-use · voice)

DVLAA (Damn Vulnerable LLM and Agent Application) is a local LLM and Agent security range for OWASP LLM Top 10 and Agent security training.

🛡️ 大模型攻防渗透测试靶场 · 提示注入CTF / OWASP LLM Top10 / 脆弱Agent / 资料聚合。目标模型可切换：DeepSeek直连、OpenRouter中转站(国产+国外十余款小模型)、本地Ollama(DeepSeek-R1 8B离线)。一键本地部署。

AI model health monitor for LLM apps – runtime checks for drift, hallucination risk, latency, and JSON/format quality on any OpenAI, Anthropic, or local client.

Security monitoring middleware for IChatClient (Microsoft.Extensions.AI). 55 detectors for prompt injection, hallucination, PII leakage, and operational anomalies. Intervention engine, embedded dashboard, audit forwarders to Azure Sentinel + OpenTelemetry. Drop-in middleware for any LLM client.

Türkçe yapay zeka güvenliği kaynakları — prompt injection, jailbreak, red teaming, guardrail, ajan/MCP ve RAG güvenliği. Her kaynağın yanında neden listelendiğini anlatan Türkçe açıklama; ölü linkler CI ile taranıyor. · Turkish AI/LLM security resource list

A credential is not authorisation. ACP moves the authorise/refuse decision for AI agent actions outside the model, where prompt injection cannot reach it. Specification, Dafny proofs, and a reference implementation — every claim replays locally.

OWASP LLM Top 10 (2025) Kapsamlı Türkçe Rehber — en kritik 10 LLM güvenlik riski, saldırı senaryoları ve savunma stratejileri. Türkçe LLM Security yazı serisinin 3. bölümü.

OmniRed: Multi-AI offensive security skills library for Claude, ChatGPT, Gemini & Microsoft Copilot — with unique MCP, LLM-pipeline, and AI-native attack categories. By Sunil Gentyala, Independent Researcher.

Agentic multi-agent framework for autonomous LLM vulnerability discovery: CVSSv4.0 scoring, 8-framework compliance mapping, SARIF output. OWASP LLM Top 10 (2025).

AWS-native AI Security Posture Management for AI agents (Observe • Govern • Defend). Discovery, OWASP/NIST/MITRE posture rules, runtime detection & Bedrock Guardrails enforcement, and AIDR integrations (Security Hub, GuardDuty) — open source, in-account, extensible.

🛡️ Open-source AI security scanner & LLM red-teaming platform. Test LLM APIs, chatbots, agents, MCP servers & RAG for prompt injection, jailbreaks, data leaks & unsafe tool use — with OWASP LLM Top 10 mapping and plain-English, audit-ready reports.

OWASP Top 10 for LLM Applications × MITRE ATLAS — a cross-version reference (2025 → 2026). Bilingual EN/TR.

Yang Code Review (YCR) - A lightweight GitHub Action for AI-powered, OWASP-aligned code review with your own LLM provider.

Hands-on AI security research covering prompt injection, LLM red-teaming, AI supply chain risks, and data poisoning. Features a Groq-powered interactive simulator (16 scenarios, 5 live LLM chats) and a NeMo Guardrails + Colang production demo. All 25 TryHackMe AI1 rooms mapped to OWASP LLM Top 10 and MITRE ATLAS.

Living aggregated LLM red-team corpus + cross-framework taxonomy (OWASP LLM Top 10, MITRE ATLAS, NIST AI RMF, EU AI Act, ISO 42001, KISA). CC-BY-4.0 (data) + MIT (code). Quarterly refresh.

Generate adversarial prompts and evaluate an offline detector for LLM red-team experiments.

A working technical blueprint for EU AI Act compliance — for Providers and Deployers. 79 controls mapped to 6 frameworks, threat model included.

AI / LLM security study notes — OWASP LLM & Agentic Top 10 crosswalks, paper notes (IPI / MSJ / Crescendo), and defence design notes.

Self-contained LLM red-team lab: deliberately-vulnerable app + garak/promptfoo/PyRIT harness mapped to OWASP LLM Top 10 (lab-only).