{
 "public_id": "AIR-2026-0142",
 "incident_id": "2026-06-openclaw-hackmyclaw-challenge",
 "date_disclosed": "2026-06-25",
 "date_occurred": null,
 "title": "What happened after 2,000 people tried to hack my AI assistant: 6,000+ injection attempts against an OpenClaw agent, zero secret leaks",
 "class": "research_demo",
 "attack_surface": "autonomous_ops",
 "initial_vector": "direct_and_indirect_prompt_injection_via_email",
 "agent_capability_abused": [
  "email_read",
  "email_send",
  "file_read"
 ],
 "target_product": "Fiu, an OpenClaw assistant on Claude Opus 4.6",
 "target_vendor": "OpenClaw / Anthropic",
 "autonomy_level": "semi_autonomous",
 "impact_realized": false,
 "impact_type": [],
 "blast_radius": "single researcher-operated instance (hackmyclaw.com)",
 "cve": [],
 "taxonomy": {
  "owasp_asi": [
   "ASI01",
   "ASI03"
  ],
  "mitre_atlas": [],
  "labels": [
   "negative_result",
   "crowdsourced_red_team",
   "system_prompt_defense",
   "secrets_in_files"
  ]
 },
 "defense_present": "system-prompt instructions prohibiting credential disclosure, file modification, code execution and data exfiltration",
 "defense_bypassed": false,
 "fix_shipped": "n/a",
 "primary_sources": [
  "https://www.fernandoi.cl/posts/hackmyclaw/"
 ],
 "verified_quote": "The secrets never leaked. No attacker managed to make Fiu send an unauthorized reply.",
 "confidence": "high",
 "notes": "Fernando Irarrazaval, independent. Included deliberately as a NEGATIVE result to balance a corpus that structurally over-samples successful attacks: 2,000+ participants and 6,000+ emails failed to extract a secrets.env file. Caveats the author himself raises: contamination from batch processing, Gmail suspension, $500+ API cost, and a narrow objective (one file, one channel). Attack styles seen: admin/incident-responder impersonation, urgency framing, reverse psychology, multi-language, authority spoofing. Surfaced via Simon Willison's prompt-injection tag.",
 "risk_category": "security_fraud",
 "risk_secondary": [
  "privacy"
 ],
 "risk_subcategory": "Prompt Injection",
 "risk_sub_secondary": [
  "Credential & Secret Theft"
 ],
 "ai_entity": "generative",
 "agency": "agent_acted",
 "agency_debatable": true,
 "agency_reason": "the agent read and sent mail throughout, but nothing leaked",
 "vector_family": "indirect_prompt_injection",
 "vector_detail": "direct_and_indirect_prompt_injection_via_email",
 "guardrail_kind": [
  "prompt_instruction"
 ],
 "guardrail_outcome": "silent",
 "xrefs": {
  "aiid": [],
  "atlas": [],
  "cve": [],
  "ghsa": []
 },
 "primary_asi": "ASI01",
 "agent_framework": "openclaw",
 "model_provider": "anthropic",
 "model_name": "",
 "detection_mechanism": "external_researcher",
 "tool_access": "unknown",
 "financial_loss_band": "unknown",
 "eai_category": "confused_deputy",
 "eai_domain": "access_control_and_permissions",
 "eai_secondary": [],
 "eai_reason": "impact rule: confused_deputy",
 "asi_curated": [
  "ASI01"
 ],
 "asi_scope": "in_scope",
 "asi_provenance": "curated_plus_rule",
 "asi_reason": "curated: ASI01 | added ASI03 (eai_category=confused_deputy)",
 "submitted_by": "EAI Internal"
}