{
 "id": "AIR-2026-0350",
 "slug": "rce-vulnerabilities-in-llm-integrated-applications-via-promp",
 "title": "RCE Vulnerabilities in LLM-Integrated Applications via Prompt Injection",
 "summary": "Researchers identified 20 remote code execution vulnerabilities across 11 different LLM frameworks. The study showed that applications built with these frameworks could be exploited using prompt injection.",
 "type": "attack_on_ai",
 "categories": [
  "prompt_injection",
  "infra_vuln"
 ],
 "severity": 1,
 "targets": [
  "framework_or_tooling"
 ],
 "products": [
  "LangChain",
  "LlamaIndex",
  "Langflow",
  "Auto-GPT"
 ],
 "incident_date": "2025-02-27",
 "incident_date_precision": "day",
 "first_reported": "2026-03-31",
 "sources": [
  {
   "url": "https://atlas.mitre.org/studies/AML.CS0052",
   "publisher": "MITRE ATLAS",
   "title": "LLMSmith: RCE Vulnerabilities in LLM-Integrated Applications",
   "published_at": "2026-03-31"
  },
  {
   "url": "https://arxiv.org/abs/2309.02926",
   "publisher": "arXiv",
   "title": "Demystifying RCE Vulnerabilities in LLM-Integrated Apps",
   "published_at": "2026-03-31"
  },
  {
   "url": "https://sites.google.com/view/llmsmith",
   "publisher": "sites.google.com",
   "title": "LLMSmith Website",
   "published_at": "2026-03-31"
  }
 ],
 "source_count": 3,
 "confidence": 0.95,
 "status": "published",
 "origin": "seed_atlas",
 "tags": [
  "rce",
  "frameworks",
  "prompt-injection"
 ],
 "created_at": "2026-10-18T06:55:08.239Z",
 "updated_at": "2026-10-18T06:55:08.239Z"
}