{
  "id": "sysdig-captures-llm-agent-driving-marimo-2026-05",
  "name": "Sysdig captures LLM agent driving a marimo exploit to an internal database dump in four pivots",
  "date_reported": "2026-05-26",
  "researcher": "unknown",
  "org": "Sysdig Threat Research Team",
  "sources": [
    {
      "url": "https://sysdig.com/blog/ai-agent-at-the-wheel-how-an-attacker-used-llms-to-move-from-a-cve-to-an-internal-database-in-4-pivots/",
      "publisher": "Sysdig Threat Research Team"
    },
    {
      "url": "https://labs.cloudsecurityalliance.org/research/csa-research-note-llm-agent-postexploit-marimo-20260602-csa/"
    }
  ],
  "borderline": false,
  "ai_role": "attacker_uses_ai_as_operator",
  "technique_summary": "Sysdig documented what it calls its first captured LLM-agent-driven intrusion, chaining a marimo RCE to AWS secrets, SSH and a PostgreSQL credential-table dump in about an hour.",
  "initial_access": "unknown",
  "ai_tooling_and_models": [
    "Unspecified LLM agent used by attacker",
    "victim ran marimo Python notebook"
  ],
  "other_tooling": [],
  "victims_and_sectors": "Technology and AI companies; Multiple sectors",
  "regions": [
    "unknown"
  ],
  "region_notes": "Target geography was not captured by the Notion feed. Actor attribution is not used as a victim location.",
  "attribution_or_cluster": "Unnamed (source IP in Indonesia)",
  "scale_and_metrics": "unknown",
  "prevention_angle": "unknown",
  "related_magicsword_projects": [],
  "source_notes": "Imported from the AI Threat Watch Notion feed (an automated daily scan); not yet editorially reviewed. Feed severity: High. Feed novelty: New TTP.",
  "tags": [
    "notion-import",
    "agentic-operations",
    "vuln-discovery-exploit"
  ],
  "category": "adversary-use-of-ai",
  "ai_involvement_confidence": "vendor-asserted"
}