{
  "id": "malicious-openclaw-skills-2026",
  "name": "Multiple campaigns distributing malicious OpenClaw skills",
  "date_reported": "2026-06-23",
  "researcher": "Koi Security, Cyera Research",
  "org": "Koi Security, Cyera",
  "sources": [
    {
      "url": "https://koi.security/blog/clawhavoc-malicious-openclaw-skills-amos",
      "title": "Koi: ClawHavoc Campaign Delivers AMOS via Malicious OpenClaw Skills"
    },
    {
      "url": "https://www.cyera.com/research/poisoned-skills-target-developer-credentials",
      "title": "Cyera: Poisoned Skills Target Developer Credentials",
      "publisher": "Cyera Research"
    }
  ],
  "borderline": false,
  "borderline_note": null,
  "ai_role": "ai_credential_theft",
  "ai_role_secondary": "ai_tool_as_lure",
  "ai_role_detail": "Multiple adversary campaigns distributing malicious OpenClaw skills via fake repos and skill registries to steal AI credentials and deliver malware",
  "technique_summary": "Koi reported ClawHavoc skills distributed through fake repositories and a rogue registry to deliver infostealers and harvest developer credentials. Cyera later reported poisoned skills using fake prerequisite lures, OS-specific installation instructions, external paste relays, and inflated README droppers. Detection should verify skill provenance, inspect companion files rather than only SKILL.md, and monitor agent-initiated shell and network activity.",
  "initial_access": "User installs malicious OpenClaw skill from fake GitHub repo, typosquat, or rogue skill registry (openclaw-skills[.]com)",
  "ai_tooling_and_models": [
    "OpenClaw AI skill framework",
    "Anthropic Claude API",
    "OpenAI API"
  ],
  "other_tooling": [
    "AMOS infostealer (macOS)",
    "Raccoon Stealer (Windows)",
    "Node.js credential harvesters",
    "Reverse shells",
    "openclaw-skills[.]com rogue registry"
  ],
  "victims_and_sectors": "AI developers installing untrusted OpenClaw skills",
  "regions": [
    "Multiple regions"
  ],
  "attribution_or_cluster": "ClawHavoc campaign and related poisoned-skill activity; actor unknown",
  "scale_and_metrics": "Multiple malicious skills and associated payload infrastructure reported during 2026",
  "prevention_angle": "Only install OpenClaw skills from trusted sources; verify GitHub repo authenticity (org, stars, commit history); review skill manifest permissions before install; use allowlisted skill sources; monitor ~/.openclaw/skills/ for unexpected additions; implement endpoint security detecting AMOS/Raccoon; educate users on skill installation risks",
  "related_magicsword_projects": [],
  "source_notes": "Koi Security ClawHavoc analysis and Cyera poisoned-skill research. Detection should monitor skill installation, unknown source repositories, unexpected network egress, and credential-file access.",
  "last_reviewed": "2026-11-04",
  "iocs": {
    "domains": [
      "openclaw-skills.com",
      "socifiapp.com",
      "custom-registry.com",
      "openclawskills.net",
      "install.app-distribution.net",
      "openclawcli.vercel.app",
      "download.setup-service.com",
      "laosji.net",
      "letssendit.fun",
      "monster-secure.site",
      "vorderingenonline.com"
    ],
    "ips": [
      "91.92.242.30",
      "5.42.65.10",
      "45.137.65.112",
      "185.234.72.19",
      "2.26.75.16"
    ],
    "hashes": [],
    "urls": [
      "https://glot.io/snippets/hfdxv8uyaf",
      "https://glot.io/snippets/hfd3x9ueu5",
      "https://rentry.co/openclaw-code"
    ]
  },
  "atlas_techniques": [
    {
      "id": "AML.T0000",
      "rationale": "Multiple campaigns delivering malicious AI skill framework extensions to steal credentials and deploy malware on AI developer machines",
      "source_urls": [
        "https://koi.security/blog/clawhavoc-malicious-openclaw-skills-amos"
      ]
    }
  ],
  "attack_flow": {
    "nodes": [
      {
        "id": "distribute-skills",
        "title": "Distribute malicious skills",
        "description": "ClawHavoc: 23 skills on fake GitHub repos + openclaw-skills[.]com registry. Unit 42: 5 skills via typosquat repos + SEO-poisoned blogs. Skills named to appear legitimate (productivity-boost, code-optimizer, security-scanner, api-connector-pro, github-integration-plus)",
        "kind": "access",
        "source_urls": [
          "https://koi.security/blog/clawhavoc-malicious-openclaw-skills-amos"
        ]
      },
      {
        "id": "user-installs",
        "title": "User installs skill",
        "description": "User finds malicious skill via search, fake registry, or social engineering; installs to ~/.openclaw/skills/ via OpenClaw CLI or UI; skill auto-loads on next OpenClaw startup",
        "kind": "ai",
        "source_urls": [
          "https://koi.security/blog/clawhavoc-malicious-openclaw-skills-amos"
        ]
      },
      {
        "id": "execute-payload",
        "title": "Execute malware and harvest",
        "description": "ClawHavoc: deploys AMOS (macOS) or Raccoon Stealer (Windows); harvests Anthropic keys, OpenAI keys, GitHub tokens, AWS creds, crypto wallets, browser cookies. Unit 42: Node.js credential harvester + reverse shell; targets same credential types",
        "kind": "execution",
        "source_urls": [
          "https://koi.security/blog/clawhavoc-malicious-openclaw-skills-amos"
        ]
      },
      {
        "id": "exfiltrate-persist",
        "title": "Exfiltrate and maintain access",
        "description": "Credentials POST to attacker C2; reverse shells provide persistent access; 6,900+ affected installations across both campaigns (ClawHavoc 4,800, Unit 42 2,100)",
        "kind": "impact",
        "source_urls": [
          "https://koi.security/blog/clawhavoc-malicious-openclaw-skills-amos"
        ]
      }
    ],
    "edges": [
      {
        "source": "distribute-skills",
        "target": "user-installs"
      },
      {
        "source": "user-installs",
        "target": "execute-payload"
      },
      {
        "source": "execute-payload",
        "target": "exfiltrate-persist"
      }
    ]
  },
  "category": "attacks-on-ai",
  "attribution": {
    "actor_name": "ClawHavoc",
    "country": "Unknown",
    "motivation": "financially-motivated",
    "attributed_by": "Koi Security",
    "confidence": "medium"
  },
  "ai_involvement_confidence": "confirmed",
  "detections": [
    {
      "name": "Linux Base64 Encoded Pipe to Shell",
      "format": "sigma",
      "url": "https://github.com/SigmaHQ/sigma/blob/master/rules/linux/process_creation/proc_creation_lnx_base64_execution.yml",
      "notes": "Detects base64 -D | bash patterns; directly applicable to Cyera-documented AMOS macOS delivery via base64-decoded pipes"
    },
    {
      "name": "Keychain Password Retrieval via Command Line",
      "format": "elastic",
      "url": "https://github.com/elastic/detection-rules/blob/main/rules/macos/credential_access_keychain_pwd_retrieval_security_cmd.toml",
      "notes": "Detects macOS keychain credential access; relevant for AMOS infostealer credential theft on macOS"
    },
    {
      "name": "Linux Shell Pipe to Shell",
      "format": "sigma",
      "url": "https://github.com/SigmaHQ/sigma/blob/master/rules/linux/process_creation/proc_creation_lnx_susp_pipe_shell.yml",
      "notes": "Detects suspicious shell pipes; applicable to AMOS delivery mechanisms using curl | bash patterns"
    },
    {
      "name": "OpenClaw skill supply-chain monitoring",
      "format": "guidance",
      "notes": "Monitor ~/.openclaw/skills/ for unexpected installations; alert on skills from unknown publishers (hightower6eu, Demerzels-lab, LJT-520, dexiaong, Ddoy233); detect fake openclaw-agent prerequisite lures; scan SKILL.md for inflated README droppers (22 MB+); flag Skills Directory A/B grades on skills with external paste relay references (glot.io, rentry.co); monitor outbound connections to AMOS C2 91.92.242.30 and secondary IPs; detect base64 -D | bash with decoy echo swcdn.apple.com; block domains socifiapp.com, custom-registry.com, openclawskills.net, laosji.net; audit skills with OS-split instructions for macOS (bash delivery) vs Windows (ZIP/EXE)"
    }
  ],
  "timeline": {
    "first_seen": "2026-02-01",
    "last_updated": "2026-11-06",
    "updates": [
      {
        "date": "2026-02-01",
        "source": "Koi Security",
        "summary": "ClawHavoc campaign discovered delivering AMOS via 23 malicious skills"
      },
      {
        "date": "2026-06-23",
        "source": "Unit 42",
        "summary": "Five malicious skills with Node.js credential harvesters and reverse shells identified"
      },
      {
        "date": "2026-11-03",
        "source": "Cyera Research",
        "summary": "Analysis of 1M+ skills reveals 516 AMOS-linked samples, 70 confirmed-malicious active in Sep 2026; all received A/B Skills Directory grades despite malicious payloads; primary C2 91.92.242.30 in 83.7% of references; fake openclaw-agent lures, OS-split delivery (base64 bash on macOS, ZIP/EXE on Windows), paste relay infrastructure"
      }
    ]
  }
}