CVE-2026-27009
📋 TL;DR
OpenClaw Control UI had a stored XSS vulnerability where attacker-controlled JavaScript could execute in the Control UI origin. This affected OpenClaw versions before 2026.2.15. Attackers could craft malicious assistant identity data containing script-breaking payloads.
💻 Affected Systems
- OpenClaw
📦 What is this software?
Openclaw by Openclaw
⚠️ Risk & Real-World Impact
Worst Case
Complete compromise of Control UI session leading to administrative account takeover, data exfiltration, or lateral movement within the environment.
Likely Case
Session hijacking, credential theft, or unauthorized actions performed within the Control UI by authenticated users.
If Mitigated
No impact with proper Content Security Policy and updated version.
🎯 Exploit Status
Exploitation requires ability to inject malicious assistant identity data. Likely requires some level of access to modify configuration.
🛠️ Fix & Mitigation
✅ Official Fix
Patch Version: 2026.2.15
Vendor Advisory: https://github.com/openclaw/openclaw/security/advisories/GHSA-37gc-85xm-2ww6
Restart Required: Yes
Instructions:
1. Backup current configuration. 2. Update OpenClaw to version 2026.2.15 or later. 3. Restart OpenClaw service. 4. Verify Content Security Policy headers are present.
🔧 Temporary Workarounds
Implement CSP Header
allAdd Content Security Policy header to block inline scripts
Add 'Content-Security-Policy: script-src \'self\'' to web server configuration
Sanitize Assistant Data
allImplement input validation and output encoding for assistant identity fields
Implement HTML entity encoding for all user-controlled data rendered in script contexts
🧯 If You Can't Patch
- Implement strict Content Security Policy with 'script-src \'self\'' and no inline scripts
- Disable or restrict access to assistant identity modification functionality
🔍 How to Verify
Check if Vulnerable:
Check OpenClaw version. If version < 2026.2.15, check if assistant identity data can contain script-breaking payloads.Check Version:
openclaw --version or check package managerVerify Fix Applied:
Verify version is 2026.2.15+, check for Content Security Policy headers, confirm no inline script injection in rendered pages.📡 Detection & Monitoring
Log Indicators:
- Unusual assistant identity modifications
- JavaScript execution errors in Control UI logs
- Multiple failed login attempts from Control UI
Network Indicators:
- Unexpected outbound connections from Control UI host
- Unusual HTTP requests to external domains
SIEM Query:
source="openclaw.log" AND ("</script>" OR "javascript:" OR "onerror=" OR "onload=")🔗 References
- https://github.com/openclaw/openclaw/commit/3b4096e02e7e335f99f5986ec1bd566e90b14a7e
- https://github.com/openclaw/openclaw/commit/adc818db4a4b3b8d663e7674ef20436947514e1b
- https://github.com/openclaw/openclaw/releases/tag/v2026.2.15
- https://github.com/openclaw/openclaw/security/advisories/GHSA-37gc-85xm-2ww6
