# ClawHub Marketplace Skills Expose OpenClaw AI Agents to RCE, Data Theft, and Supply-Chain Backdoors (CVE-2026-25253)

> ClawHub, the official skill marketplace for the open-source OpenClaw AI agent platform, grew from under 2,000 to over 50,000 skills between January and April 2026 with minimal security review, enabling a documented supply-chain crisis: the ClawHavoc campaign (1,184 malicious skills, 247,000 installs, $2.3M in stolen cryptocurrency via Atomic macOS Stealer), a critical one-click RCE (CVE-2026-25253, CVSS 8.8) exploiting unvalidated gatewayUrl WebSocket token exfiltration, a Python-pickle-deserialization RCE backdoor disguised as a "distributed state recovery tool," and a download-ranking manipulation flaw that let malicious skills auto-install via autonomous agent tool selection.

- **Published:** 2026-07-11T00:00:00Z
- **Last reviewed:** 2026-07-11T00:00:00Z
- **Canonical:** https://intel.threadlinqs.com/threat/TL-2026-1212
- **Card image:** https://intel.threadlinqs.com/og/threat/threat-intel-clawhavoc-clawhub-skill-campaign-tl-2026-1212.webp?v=13684233 (Threadlinqs threat intel: ClawHavoc OpenClaw ClawHub Skills RCE: Detection & IOCs (TL-2026-1212) — High severity, 23 IOCs, 9 detection rules.)
- **ID:** TL-2026-1212
- **Severity:** HIGH (CVSS 8.8)
- **Category:** SUPPLY_CHAIN
- **Status:** ACTIVE
- **Detections:** 9 (1 shown free below) · **IOCs:** 23 (3 shown free below)
- **CVEs:** CVE-2026-25253

## Description

OpenClaw is an open-source AI agent framework that grants installed "skills" full system permissions — file read/write, network access, and shell command execution — immediately upon installation, with no sandboxing or per-permission scoping. ClawHub, its official skill marketplace, scaled from fewer than 2,000 skills in January 2026 to over 50,000 by April 2026, a 25x expansion in under 90 days, dramatically outpacing any vetting capacity.

This explosive, unvetted growth produced four distinct but related security failures documented across independent researchers between February and June 2026:

1. **ClawHavoc supply-chain campaign**: Beginning around January 2026, threat actors compromised or created developer accounts on ClawHub to publish malicious skills. Koi Security's Oren Yomtov first disclosed 341 malicious skills out of 2,857 scanned on February 1, 2026; a follow-up scan on February 16 found 824 malicious skills in a registry that had grown to 10,700+; Antiy Labs ultimately catalogued 1,184 malicious skills published historically to ClawHub via 12 compromised developer accounts, resulting in 247,000 confirmed installations and $2.3 million in stolen cryptocurrency. Payloads used typosquatted skill names ("Google Assistant Pro," "YouTube Summarize Pro," wallet/Polymarket/YouTube utilities) to deliver the Atomic macOS Stealer (AMOS) trojan, embedded in shell scripts and later pivoted to skill-page comments disguised as "update service" instructions after SKILL.md-focused scanning caught the initial vector.

2. **CVE-2026-25253 — one-click RCE via auth-token exfiltration**: OpenClaw's Control UI (affecting all versions before 2026.1.29) reads a `gatewayUrl` value from a query string and automatically establishes a WebSocket connection without user confirmation, transmitting the operator's authentication token to whatever host the URL specifies. Because OpenClaw operates with full file/shell access, token theft grants an attacker complete control of the victim's machine with a single malicious link click. Public disclosure was February 3, 2026; contemporaneous internet-wide scanning found over 40,000 exposed OpenClaw instances, 63% assessed as remotely exploitable. The vendor's initial fix (2026.1.29) added a confirmation prompt when gatewayUrl changes.

3. **Pickle-deserialization RCE backdoor**: A skill masquerading as a "distributed state recovery tool," complete with legitimate-looking architecture documentation, fetched a remote payload from attacker C2, decoded it through a chained 12-method encoding scheme (Base64, ROT13, Morse, and others), and executed it via unsafe Python `pickle.loads()` deserialization. Tencent's Zhuque Lab AIG platform flagged the combination of remote fetch, chained encoding, and unsafe deserialization as the detection signature; the skill's declared networking/Python-runtime permissions appeared superficially legitimate for its stated purpose.

4. **Download-ranking manipulation → autonomous auto-install**: Silverfort researchers found an unauthenticated, unrate-limited public RPC endpoint (`download.increment`) that let any attacker with a skill ID and deployment URL arbitrarily inflate a skill's download count. Using a proof-of-concept "Outlook Graph Integration" skill with a hidden data-exfiltration payload (collecting the victim's username and FQDN to an attacker-controlled server), the researchers drove the skill to the #1 ranking in its category, producing 3,900 autonomous executions within 6 days across 50+ cities, including at several public companies — because OpenClaw agents autonomously prioritize installing the highest-ranked skill when selecting tools. Reported March 16, 2026; ClawHub shipped a fix within ~24 hours.

Broader corpus analysis by Tencent Zhuque Lab (scanning ~50,000 skills) found 74.6% declare network-request permissions and roughly 25% have file read/write capability, with 246,378 distinct external URLs (29,196 unique domains) referenced across skill code, several overlapping with known C2 infrastructure. A 15,427-developer base is heavily concentrated: the top 20 developers account for 5,422 skills (12.9% of the corpus), and one account alone published 955 skills in 90 days — a pattern consistent with automated batch-generation rather than organic development, strongly correlated with the malicious clusters. Shanghai Jiao Tong University's SkillProbe team independently found over 90% of highly-downloaded skills failed rigorous security audits. Following disclosure, ClawHub integrated VirusTotal screening (Feb 2026) and partnered with NVIDIA for tooling/documentation analysis (June 1, 2026), though Unit 42 documented at least five distinct malicious skills evading these controls between February and May 2026 via 22MB README padding (exceeding scanner size thresholds), paste-site redirect indirection (allowing payload updates without republishing), and runtime affiliate-link/front-running injection schemes.

## MITRE ATT&CK

- T1195 Supply Chain Compromise
- T1566 Phishing
- T1199 Trusted Relationship
- T1203 Exploitation for Client Execution
- T1059 Command and Scripting Interpreter
- T1059.006 Python
- T1059.004 Unix Shell
- T1204.001 Malicious Link
- T1211 Exploitation for Stealth
- T1547.013 XDG Autostart Entries
- T1548 Abuse Elevation Control Mechanism
- T1036 Masquerading
- T1027 Obfuscated Files or Information
- T1140 Deobfuscate/Decode Files or Information
- T1070.004 File Deletion
- T1528 Steal Application Access Token
- T1552.001 Credentials In Files
- T1539 Steal Web Session Cookie
- T1083 File and Directory Discovery
- T1082 System Information Discovery
- T1005 Data from Local System
- T1213 Data from Information Repositories
- T1071.001 Web Protocols
- T1571 Non-Standard Port
- T1102 Web Service
- T1105 Ingress Tool Transfer
- T1205.002 Socket Filters
- T1041 Exfiltration Over C2 Channel
- T1657 Financial Theft
- T1586 Compromise Accounts
- T1583.001 Domains

## Free detection (Splunk SPL, Microsoft KQL, Sigma)

**OpenClaw Control UI gatewayUrl Auto-WebSocket Auth-Token Exfiltration (CVE-2026-25253)** (`TL-2026-1212-SPL-001`)

Detects the OpenClaw Control UI process establishing an outbound network connection immediately after a gatewayUrl query-string value changes, without a corresponding user-confirmation event — the core exploitation mechanism of CVE-2026-25253's one-click RCE via auth-token theft.

- **Languages:** Splunk SPL, Microsoft KQL, Sigma
- **ATT&CK:** T1528, T1203, T1204.001
- **Data sources:** sysmon, windows_security, endpoint_dns_network
- **Severity:** critical
- **Confidence:** high

### Splunk SPL

```spl
index=endpoint sourcetype="XmlWinEventLog:Microsoft-Windows-Sysmon/Operational" (EventCode=1 OR EventCode=3) Image="*openclaw-control-ui.exe"
| eval has_gateway_param=if(match(CommandLine, "(?i)gatewayurl="), 1, 0)
| eval has_confirm_flag=if(match(CommandLine, "(?i)--gatewayurl-confirmed"), 1, 0)
| where has_gateway_param=1 AND has_confirm_flag=0
| stats count min(_time) as first_seen max(_time) as last_seen by host, User, CommandLine, DestinationIp, DestinationPort
| where count > 0
| convert ctime(first_seen) ctime(last_seen)
| table first_seen, last_seen, host, User, CommandLine, DestinationIp, DestinationPort, count
```

### Microsoft KQL

```kql
DeviceProcessEvents
| where Timestamp > ago(24h)
| where FolderPath has "openclaw-control-ui.exe" or ProcessCommandLine has_cs "gatewayUrl"
| where ProcessCommandLine has "gatewayUrl" and not(ProcessCommandLine has "--gatewayUrl-confirmed")
| join kind=inner (
    DeviceNetworkEvents
    | where Timestamp > ago(24h)
    | where InitiatingProcessFileName =~ "openclaw-control-ui.exe"
) on DeviceId, $left.ProcessId == $right.InitiatingProcessId
| project Timestamp, DeviceName, AccountName, ProcessCommandLine, RemoteIP, RemoteUrl, RemotePort
| order by Timestamp desc
```

### Sigma

```yaml
title: OpenClaw Control UI Outbound Connection Following gatewayUrl Change Without Confirmation
id: 6a1b2c3d-4e5f-4711-8a12-tl20261212001
status: experimental
description: Detects the OpenClaw Control UI auto-establishing an outbound WebSocket connection right after a gatewayUrl query-string parameter changes, with no operator confirmation event — the exploitation signature of CVE-2026-25253.
references:
    - https://github.com/advisories/GHSA-g8p2-7wf7-98mq
    - https://nvd.nist.gov/vuln/detail/CVE-2026-25253
author: Threadlinqs Team
date: 2026-07-11
tags:
    - attack.credential-access
    - attack.initial-access
    - attack.t1528
    - attack.t1203
    - cve.2026.25253
    - attack.tool.openclaw_control_ui
logsource:
    category: network_connection
    product: windows
detection:
    selection_process:
        Image|endswith: '\\openclaw-control-ui.exe'
        Initiated: 'true'
    selection_param:
        CommandLine|contains: 'gatewayUrl'
    filter_confirmed:
        CommandLine|contains: '--gatewayUrl-confirmed'
    condition: selection_process and selection_param and not filter_confirmed
falsepositives:
    - Legitimate operator-approved gateway reconfiguration performed through the patched confirmation-prompt flow (should carry the confirmation flag)
level: high
```

**Known false positives:**

- Legitimate operator-approved gateway reconfiguration performed on OpenClaw 2026.1.29+ where the confirmation-prompt flag is correctly logged

**Tuning notes:** Baseline your organization's patched OpenClaw version rollout; suppress alerts once fleet-wide upgrade to 2026.1.29+ is confirmed via version-inventory telemetry, and require the --gatewayUrl-confirmed flag as a hard allowlist condition rather than relying on host age.

### 8 more detection rules (Splunk SPL, Microsoft KQL, Sigma), unlock with TL Creds

- 'Distributed State Recovery Tool' Chained-Encoding Pickle-Deserialization Backdoor (locked: unlock with TL Creds)
- Oversized SKILL.md/README Padding Consistent with ClawScan/VirusTotal Evasion (locked: unlock with TL Creds)
- Skill Permission Declaration Disproportionate to Stated Function (Over-Broad Grant Detection) (locked: unlock with TL Creds)
- Unauthenticated ClawHub download.increment RPC Abuse Enabling Ranking-Manipulation Auto-Install (locked: unlock with TL Creds)
- Autonomous Agent Auto-Install of Newly Top-Ranked Skill Within Anomalously Short Time-to-Rank (locked: unlock with TL Creds)
- OpenClaw Skill Process Beaconing to Known ClawHavoc/Pickle-Backdoor C2 Infrastructure (locked: unlock with TL Creds)
- ClawHub Typosquatted Skill curl-pipe-bash Install Delivery Chain (locked: unlock with TL Creds)
- AMOS/cluw Infostealer Post-Install Keychain and Credential File Collection (locked: unlock with TL Creds)

Sign up free to unlock the detection rules: https://threadlinqs.com/?modal=signup

## Sample indicators (3 of 23)

Defanged, TLP:CLEAR.

- IP address: `2.26.75[.]16`
- Domain: `laosji[.]net`
- URL: `hxxps://rentry[.]co/openclaw-code`

Full set of 23 indicators unlocks with TL Creds: https://threadlinqs.com/tl-creds

## Sources

- [ClawHub Skills Expose AI Agents to Remote Control Backdoors and Data Theft Attacks](https://cybersecuritynews.com/clawhub-skills-expose-ai-agents/)
- [Malicious OpenClaw Skills Used to Distribute Atomic MacOS Stealer](https://www.trendmicro.com/en_us/research/26/b/openclaw-skills-used-to-distribute-atomic-macos-stealer.html)
- [OpenClaw's Skill Marketplace and the Emerging AI Supply Chain Threat](https://unit42.paloaltonetworks.com/openclaw-ai-supply-chain-risk/)
- [Researchers Find 341 Malicious ClawHub Skills Stealing Data from OpenClaw Users](https://thehackernews.com/2026/02/researchers-find-341-malicious-clawhub.html)
- [ClawHavoc Poisons OpenClaw's ClawHub With 1,184 Malicious Skills](https://cyberpress.org/clawhavoc-poisons-openclaws-clawhub-with-1184-malicious-skills/)
- [ClawHavoc campaign: 335 malicious OpenClaw skills distribute Atomic macOS Stealer via ClawHub](https://armorerlabs.com/threat-intel/openclaw-clawhavoc-campaign-amos-2026-02)
- [We Scanned 50,000 Skills: The Threat Persists — Tencent Zhuque Lab](https://matrix.tencent.com/2026/04/24/we-scanned-50000-skills-the-threat-persists)
- [ClawHub vulnerability puts malicious skill at #1 — Silverfort](https://www.silverfort.com/blog/clawhub-vulnerability-enables-attackers-to-manipulate-rankings-to-become-the-number-one-skill/)
- [ClawHub Vulnerability Lets Attackers Manipulate Rankings to Reach #1](https://cyberpress.org/clawhub-vulnerability/)
- [CVE-2026-25253 detail - NVD](https://nvd.nist.gov/vuln/detail/CVE-2026-25253)
- [OpenClaw/Clawdbot has 1-Click RCE via Authentication Token Exfiltration From gatewayUrl](https://github.com/advisories/GHSA-g8p2-7wf7-98mq)
- [OpenClaw Auth Token Theft Leading to RCE: CVE-2026-25253](https://www.sonicwall.com/blog/openclaw-auth-token-theft-leading-to-rce-cve-2026-25253)
- [CVE-2026-25253: 1-Click RCE in OpenClaw Through Auth Token Exfiltration](https://socradar.io/blog/cve-2026-25253-rce-openclaw-auth-token/)
- [One Click to Compromise: Inside OpenClaw's Critical RCE Flaw — Hive Pro](https://hivepro.com/threat-advisory/one-click-to-compromise-inside-openclaws-critical-rce-flaw/)
- [GitHub - adibirzu/openclaw-security-monitor: Detects ClawHavoc, AMOS stealer, CVE-2026-25253, memory poisoning, and supply chain attacks](https://github.com/adibirzu/openclaw-security-monitor)

## Full data

The other 8 detection rules (Splunk SPL / Microsoft KQL / Sigma) unlock with TL Creds; a free account includes daily creds: https://threadlinqs.com/?modal=signup
Indicator values unlock with TL Creds, at a lower price on Red and Purple; AI agents can query rules and IOCs through the Threadlinqs MCP server (Purple): https://intel.threadlinqs.com/mcp

Canonical: https://intel.threadlinqs.com/threat/TL-2026-1212
