{
 "public_id": "AIR-2025-0173",
 "incident_id": "2025-11-servicenow-bodysnatcher",
 "date_disclosed": "2025-11-05",
 "date_occurred": null,
 "title": "BodySnatcher: unauthenticated impersonation of any ServiceNow user through Now Assist AI agent linking",
 "class": "responsible_disclosure",
 "attack_surface": "enterprise_assistant",
 "initial_vector": "authentication_bypass_static_secret",
 "agent_capability_abused": [
  "ai_agent_execution",
  "virtual_agent_api",
  "privileged_workflow_execution"
 ],
 "target_product": "ServiceNow Now Assist AI Agents / Virtual Agent API",
 "target_vendor": "ServiceNow",
 "autonomy_level": "semi_autonomous",
 "impact_realized": false,
 "impact_type": [
  "account_takeover",
  "privilege_escalation",
  "data_exfiltration"
 ],
 "blast_radius": "ServiceNow instances running affected Now Assist AI Agents and Virtual Agent API versions",
 "cve": [
  "CVE-2025-12420"
 ],
 "taxonomy": {
  "owasp_asi": [
   "ASI03"
  ],
  "mitre_atlas": [],
  "labels": [
   "auth_bypass",
   "hardcoded_secret",
   "mfa_bypass",
   "agent_identity",
   "impersonation"
  ]
 },
 "defense_present": "MFA and SSO on the ServiceNow instance",
 "defense_bypassed": true,
 "fix_shipped": "ServiceNow remediated hosted instances 2025-10-30; fixed versions Now Assist AI Agents 5.1.18 / 5.2.19 and Virtual Agent API 3.15.2 / 4.0.4",
 "primary_sources": [
  "https://appomni.com/ao-labs/bodysnatcher-agentic-ai-security-vulnerability-in-servicenow/"
 ],
 "verified_quote": "A Virtual Agent integration flaw...allowed unauthenticated attackers to impersonate any ServiceNow user using only an email address, bypassing MFA and SSO.",
 "confidence": "high",
 "notes": "AppOmni Labs. Reported to ServiceNow 2025-10-23, remediated 2025-10-30. Root cause: AI agent providers shipped with an identical hardcoded secret across every ServiceNow instance, plus auto-linking logic that trusted a supplied email address. Reported CVSS 9.3. Not a prompt-injection bug; it is an agent-identity/authentication failure in an agent platform, which is why it belongs in this corpus. ServiceNow said it had no evidence of pre-fix exploitation. date_disclosed is the public-advisory window; the exact publication day was not verified from the AppOmni page.",
 "risk_category": "security_fraud",
 "risk_secondary": [
  "privacy"
 ],
 "risk_subcategory": "Credential & Secret Theft",
 "risk_sub_secondary": [
  "Unauthorized Access & Account Takeover"
 ],
 "ai_entity": "generative",
 "agency": "ai_as_target",
 "agency_debatable": true,
 "agency_reason": "a hard-coded static secret allowed unauthenticated agent linking",
 "vector_family": "other",
 "vector_detail": "authentication_bypass_static_secret",
 "guardrail_kind": [
  "authentication"
 ],
 "guardrail_outcome": "silent",
 "xrefs": {
  "aiid": [],
  "atlas": [],
  "cve": [
   "CVE-2025-12420"
  ],
  "ghsa": []
 },
 "primary_asi": "ASI03",
 "agent_framework": "unknown",
 "model_provider": "unknown",
 "model_name": "",
 "detection_mechanism": "unknown",
 "tool_access": "unknown",
 "financial_loss_band": "unknown",
 "eai_category": "credential_theft",
 "eai_domain": "access_control_and_permissions",
 "eai_secondary": [
  "privilege_escalation",
  "data_exfiltration_channels"
 ],
 "eai_reason": "subcategory: Credential & Secret Theft",
 "asi_curated": [],
 "asi_scope": "in_scope",
 "asi_provenance": "rule",
 "asi_reason": "ASI03 (eai_category=credential_theft)",
 "submitted_by": "EAI Internal"
}