CVE-2026-53827— OpenClaw < 2026.5.2 - Credential Exposure via Model-Supplied Loopback URLs in message.action Forwarding
Quick assessment
Affected
OpenClaw OpenClaw
Exploitation
No confirmed in-the-wild exploitation; assess based on exposure
Recommended action
Check the vendor advisory and references for a fixed version. If immediate upgrade is impossible, restrict exposure and increase monitoring.
OpenClaw是OpenClaw团队开源的一个智能人工助理。 OpenClaw 2026.5.2之前版本存在服务端请求伪造漏洞，该漏洞源于message.action转发中的凭证暴露问题，可能允许模型控制的元数据使用Gateway凭证将操作有效载荷转发到攻击者提供的回环URL，导致远程攻击者通过模型控制的操作元数据提供恶意回环目标来拦截Gateway令牌和操作有效载荷。
Get alerts for future matching vulnerabilities
Log in to subscribe
I. Basic Information for CVE-2026-53827
Vulnerability Information
Have questions about the vulnerability? See if Shenlong's analysis helps!
View Shenlong Deep Dive ↗
Although we use advanced large model technology, its output may still contain inaccurate or outdated information.Shenlong tries to ensure data accuracy, but please verify and judge based on the actual situation.
Vulnerability Title
OpenClaw < 2026.5.2 - Credential Exposure via Model-Supplied Loopback URLs in message.action Forwarding
Source: CVE Program / CVE List V5
Vulnerability Description
OpenClaw before 2026.5.2 contains a credential exposure vulnerability in message.action forwarding that allows model-controlled metadata to forward action payloads with Gateway credentials to attacker-supplied loopback URLs. Remote attackers can intercept Gateway tokens and action payloads by providing malicious loopback targets through model-controlled action metadata.
Source: CVE Program / CVE List V5
CVSS Information
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N
Source: CVE Program / CVE List V5
Vulnerability Type
服务端请求伪造(SSRF)
Source: CVE Program / CVE List V5
Vulnerability Title
OpenClaw 服务端请求伪造漏洞
Source: CNNVD (China National Vulnerability Database)
Vulnerability Description
OpenClaw是OpenClaw团队开源的一个智能人工助理。 OpenClaw 2026.5.2之前版本存在服务端请求伪造漏洞，该漏洞源于message.action转发中的凭证暴露问题，可能允许模型控制的元数据使用Gateway凭证将操作有效载荷转发到攻击者提供的回环URL，导致远程攻击者通过模型控制的操作元数据提供恶意回环目标来拦截Gateway令牌和操作有效载荷。
Source: CNNVD (China National Vulnerability Database)
CVSS Information
N/A
Source: CNNVD (China National Vulnerability Database)
Vulnerability Type
N/A
Source: CNNVD (China National Vulnerability Database)
Affected Products
II. Public POCs for CVE-2026-53827
| # | POC Description | Source Link | Shenlong Link |
|---|
AI-Generated POC
Premium
No public POC found.
Login to generate AI POCIII. Intelligence Information for CVE-2026-53827
请登录查看更多情报信息。
Vendor Advisories for CVE-2026-53827 (2)
Same Patch Batch · OpenClaw · 2026-06-12 · 20 CVEs total
| CVE-2026-53838 | 9.8 CRITICAL | OpenClaw < 2026.5.27 - Node Pairing State Mutation via Reconnection |
| CVE-2026-53822 | 8.8 HIGH | OpenClaw < 2026.5.18 - Command Argument Modification via Shell Wrapper Between Approval an |
| CVE-2026-53836 | 8.8 HIGH | OpenClaw < 2026.5.12 - Allowlist Bypass via PowerShell Encoded-Command Aliases |
| CVE-2026-53821 | 8.8 HIGH | OpenClaw < 2026.5.18 - Scope Elevation in trusted-proxy Control UI WebSocket |
| CVE-2026-53828 | 8.8 HIGH | OpenClaw < 2026.5.6 - Native Command Authorization Bypass via Owner-Command Enforcement |
| CVE-2026-53831 | 8.3 HIGH | OpenClaw < 2026.5.18 - Arbitrary File Read via Shell Expansion in system.run Safe-bin Allo |
| CVE-2026-53823 | 8.1 HIGH | OpenClaw < 2026.5.3 - Privilege Escalation via Mutable Slack Display Names in allowFrom |
| CVE-2026-53829 | 8.0 HIGH | OpenClaw < 2026.5.18 - Command Truncation in Exec Approval Display |
| CVE-2026-53833 | 7.7 HIGH | QQBot for OpenClaw < 2026.4.29 - Authorization Bypass via QQBot Streaming Command |
| CVE-2026-53832 | 7.7 HIGH | OpenClaw < 2026.5.18 - Identity Header Forgery via Trusted-Proxy Configuration |
| CVE-2026-53834 | 7.5 HIGH | OpenClaw < 2026.4.27 - Authorization Bypass in QQBot Pre-dispatch Slash Commands |
| CVE-2026-53820 | 6.6 MEDIUM | OpenClaw < 2026.5.12 - Exec Denylist Bypass in Bundle MCP Loopback Session Spawn |
| CVE-2026-53825 | 6.5 MEDIUM | OpenClaw < 2026.4.7 - Arbitrary Local File Read via memory-wiki Ingest with operator.write |
| CVE-2026-53824 | 6.5 MEDIUM | Mattermost plugin for OpenClaw < 2026.4.24 - Slash Token Revocation Lag via Monitor Refres |
| CVE-2026-53830 | 6.5 MEDIUM | OpenClaw < 2026.4.22 - Webhook Secret Revocation Bypass via secrets.reload |
| CVE-2026-53839 | 6.5 MEDIUM | OpenClaw < 2026.5.7 - Hostname Prefix Matching Bypass in Trusted Retry Endpoint Validation |
| CVE-2026-53826 | 4.3 MEDIUM | OpenClaw < 2026.4.26 - Information Disclosure via Sandboxed Session Spawn |
| CVE-2026-53835 | 4.3 MEDIUM | OpenClaw < 2026.5.6 - Config-Write Enforcement Bypass in Feishu Dynamic-Agent Bindings |
| CVE-2026-53837 | 3.7 LOW | OpenClaw < 2026.5.6 - Missing Channel Type Validation in Mattermost Event Handlers |
IV. Related Vulnerabilities
Same product: OpenClaw
-
CVE-2026-102807
OpenClaw before 2026.9.4 Authorization Bypass via MCP App St -
CVE-2026-102806
OpenClaw before 2026.9.5 Sandbox Isolation Bypass via Media -
CVE-2026-100599
OpenClaw 2026.5.1 before 2026.7.1 Remote Code Execution via -
CVE-2026-100598
OpenClaw before 2026.7.1 Approval Binding Logic Error -
CVE-2026-100597
OpenClaw before 2026.7.1 Path Traversal via Filesystem Race
Same vendor: OpenClaw
-
CVE-2026-101884
OpenClaw Windows Node before 2026.7.1 Remote Code Execution -
CVE-2026-101883
OpenClaw Windows Node through 2026.9.4 SSRF via canvas.prese -
CVE-2026-101882
OpenClaw Windows Node before 2026.7.1 Remote Code Execution -
CVE-2026-101881
OpenClaw Windows Node before 2026.7.1 Denial of Service -
CVE-2026-101880
OpenClaw Windows Node before 2026.7.1 Authorization Bypass
Same weakness: CWE-918
-
CVE-2026-103530
decolua 9Router Search Endpoint ssrfGuard.js fetch server-si -
CVE-2026-102105
Kiteworks Email Protection Gateway server-side request forge -
CVE-2023-54402
iDocView SSRF via /doc/upload Endpoint Hardcoded Token -
CVE-2026-102104
Kiteworks Email Protection Gateway server-side request forge -
CVE-2026-102103
Kiteworks Email Protection Gateway server-side request forge
V. Comments for CVE-2026-53827
No comments yet