{
 "public_id": "AIR-2026-0258",
 "incident_id": "2026-09-n8n-fromai-sandbox-escape-rce",
 "date_disclosed": "2026-09-03",
 "title": "CVE-2026-85169: n8n $fromAI prototype leak lets an expression escape the sandbox and run code on the host",
 "attack_surface": "agent_framework",
 "initial_vector": "authenticated_workflow_author_expression",
 "agent_capability_abused": [
  "fromAI_placeholder_resolution",
  "expression_evaluation"
 ],
 "target_product": "n8n",
 "target_vendor": "n8n",
 "autonomy_level": "assistant",
 "impact_realized": false,
 "impact_type": [
  "remote_code_execution"
 ],
 "blast_radius": "n8n before 1.123.73, 2.35.4 and 2.36.2 for users able to build workflows",
 "cve": [
  "CVE-2026-85169"
 ],
 "fix_shipped": "n8n 1.123.73, 2.35.4 and 2.36.2",
 "primary_sources": [
  "https://github.com/n8n-io/n8n/security/advisories/GHSA-9x83-43r8-5hwc",
  "https://www.thehackerwire.com/vulnerability/CVE-2026-85169/"
 ],
 "confidence": "high",
 "notes": "GitHub advisory GHSA-9x83-43r8-5hwc was published 2026-08-19 (CVSS 4.0 score 8.7); the CVE record was published 2026-09-03, which is the date used here. The $fromAI helper, which resolves placeholders that a model fills in, returned a live host-prototype reference, letting an expression reach the Function constructor and run code in the main n8n process. Requires workflow-build privilege. The advisory is a vulnerability description; no exploitation is reported.",
 "verified_quote": "compile code running in the main n8n process, from workflow-build privilege alone.",
 "class": "responsible_disclosure",
 "date_occurred": null,
 "defense_present": null,
 "defense_bypassed": true,
 "taxonomy": {
  "owasp_asi": [
   "ASI05"
  ],
  "mitre_atlas": [],
  "labels": [
   "sandbox_escape",
   "prototype_chain",
   "ai_expression_helper"
  ]
 },
 "risk_category": "security_fraud",
 "risk_secondary": [],
 "risk_subcategory": "Remote Code Execution & Sandbox Escape",
 "risk_sub_secondary": [],
 "ai_entity": "generative",
 "agency": "ai_as_target",
 "agency_debatable": true,
 "agency_reason": "expression sandbox escape in an automation platform; contested because $fromAI is the agent-parameter mechanism",
 "vector_family": "other",
 "vector_detail": "authenticated_workflow_author_expression",
 "guardrail_kind": [
  "unknown"
 ],
 "guardrail_outcome": "silent",
 "xrefs": {
  "aiid": [],
  "atlas": [],
  "cve": [
   "CVE-2026-85169"
  ],
  "ghsa": []
 },
 "primary_asi": "ASI05",
 "agent_framework": "n8n",
 "model_provider": "unknown",
 "model_name": "",
 "detection_mechanism": "unknown",
 "tool_access": "unknown",
 "financial_loss_band": "unknown",
 "eai_category": "privilege_escalation",
 "eai_domain": "access_control_and_permissions",
 "eai_secondary": [],
 "eai_reason": "subcategory: Remote Code Execution & Sandbox Escape",
 "asi_curated": [],
 "asi_scope": "in_scope",
 "asi_provenance": "rule",
 "asi_reason": "ASI05 (impact=code_exec)",
 "submitted_by": "EAI Internal"
}