{
 "public_id": "AIR-2026-0033",
 "incident_id": "2026-02-snyk-toxicskills",
 "date_disclosed": "2026-02-05",
 "date_occurred": null,
 "title": "ToxicSkills: 1,467 flawed and 76 confirmed-malicious agent skills across ClawHub and skills.sh",
 "class": "in_the_wild",
 "attack_surface": "skill_plugin",
 "initial_vector": "malicious_marketplace_package",
 "agent_capability_abused": [
  "skill_execution",
  "prompt_context_injection",
  "shell_execution",
  "file_read",
  "credential_access",
  "network_egress"
 ],
 "target_product": "ClawHub and skills.sh agent-skill registries",
 "target_vendor": "multiple",
 "autonomy_level": "semi_autonomous",
 "impact_realized": false,
 "impact_type": [
  "credential_theft",
  "backdoor_installation",
  "data_exfiltration",
  "persistence"
 ],
 "blast_radius": "3,984 skills scanned; 36.82% (1,467) with at least one security flaw, 13.4% (534) critical; 8 malicious skills still live at publication",
 "cve": [],
 "taxonomy": {
  "owasp_asi": [
   "ASI04",
   "ASI05",
   "ASI06",
   "ASI01",
   "ASI03"
  ],
  "mitre_atlas": [],
  "labels": [
   "supply_chain",
   "malicious_skill",
   "prompt_injection",
   "registry_audit",
   "skill_marketplace",
   "prompt_injection_in_skill",
   "credential_theft",
   "no_code_signing"
  ]
 },
 "defense_present": "none - publishing requires only a SKILL.md and a one-week-old GitHub account",
 "defense_bypassed": true,
 "fix_shipped": "reported to marketplace operators; Snyk released scanning guidance",
 "primary_sources": [
  "https://snyk.io/blog/toxicskills-malicious-ai-agent-skills-clawhub/"
 ],
 "verified_quote": "The barrier to publishing a new agent skill on ClawHub? A `SKILL.md` Markdown file and a GitHub account that's one week old. No code signing. No security review.",
 "confidence": "high",
 "notes": "Snyk Labs (Beurer-Kellner, Kudrinskii, Milanta, Bonde Nielsen, Sarkar, Tal). Named malicious publishers: zaycv (clawhud, clawhub1), Aslaep123 (polymarket-traiding-bot, base-agent, bybit-agent), pepe276 (moltbookagent, publish-dist), moonshine-100rze (moltbook-lm8). 91% of confirmed malicious skills combined conventional malware with prompt injection. Also reported as: 'ToxicSkills: Snyk audit of 3,984 agent skills on ClawHub and skills.sh finds 76 confirmed malicious payloads'.",
 "is_umbrella": true,
 "umbrella_constituents": [],
 "umbrella_reason": "Population-level ecosystem audit; no individual skill from the audit is separately cataloged.",
 "risk_category": "security_fraud",
 "risk_secondary": [
  "privacy"
 ],
 "risk_subcategory": "Prompt Injection",
 "risk_sub_secondary": [
  "Supply Chain Compromise",
  "Credential & Secret Theft",
  "Persistence & Covert C2"
 ],
 "ai_entity": "generative",
 "agency": "agent_acted",
 "agency_debatable": true,
 "agency_reason": "the agents executed the malicious skills' shell payloads",
 "vector_family": "malicious_component",
 "vector_detail": "malicious_marketplace_package",
 "guardrail_kind": [
  "none"
 ],
 "guardrail_outcome": "not_applicable",
 "xrefs": {
  "aiid": [],
  "atlas": [],
  "cve": [],
  "ghsa": []
 },
 "primary_asi": "ASI04",
 "agent_framework": "clawhub_skill",
 "model_provider": "unknown",
 "model_name": "",
 "detection_mechanism": "external_researcher",
 "tool_access": "unknown",
 "financial_loss_band": "unknown",
 "eai_category": "credential_theft",
 "eai_domain": "access_control_and_permissions",
 "eai_secondary": [
  "privilege_escalation",
  "supply_chain_vulnerabilities",
  "data_exfiltration_channels"
 ],
 "eai_reason": "impact rule: credential_theft",
 "asi_curated": [
  "ASI04",
  "ASI05",
  "ASI06"
 ],
 "asi_scope": "in_scope",
 "asi_provenance": "curated_plus_rule",
 "asi_reason": "curated: ASI04,ASI05,ASI06 | added ASI01 (vector=injection), ASI03 (eai_category=credential_theft)",
 "submitted_by": "EAI Internal"
}