CVE-2026-100573

Published: Set 26, 2026 Last Modified: Set 26, 2026
ExploitDB:
Other exploit source:
Google Dorks:
MEDIUM 4,8
Source: [email protected]
Attack Vector: local
Attack Complexity: low
Privileges Required: low
User Interaction: none
Confidentiality: N/A
Integrity: N/A
Availability: N/A
LOW 3,3
Source: [email protected]
Attack Vector: local
Attack Complexity: low
Privileges Required: low
User Interaction: none
Scope: unchanged
Confidentiality: low
Integrity: none
Availability: none

Description

AI Translation Available

OpenClaw versions before 2026.8.1 contain a sandbox policy bypass vulnerability in the MCP loopback component that allows sandboxed coding-agent sessions to invoke tools explicitly denied by sandbox.tools.deny policy. Attackers can list and invoke denied tools to access data or perform actions the operator intended to exclude from the sandbox.

862

Missing Authorization

Incomplete
Abstraction: Class Structure: Simple
Common Consequences
Security Scopes Affected:
Confidentiality Integrity Access Control Availability
Potential Impacts:
Read Application Data Read Files Or Directories Modify Application Data Modify Files Or Directories Gain Privileges Or Assume Identity Bypass Protection Mechanism Dos: Crash, Exit, Or Restart Dos: Resource Consumption (Cpu) Dos: Resource Consumption (Memory) Dos: Resource Consumption (Other)
Applicable Platforms
Technologies: AI/ML, Web Server, Database Server, Not Technology-Specific
Likelihood of Exploit: High
View CWE Details
https://github.com/openclaw/openclaw/security/advisories/GH…
https://github.com/openclaw/openclaw/security/advisories/GHSA-5m4g-88rg-69pj
https://www.vulncheck.com/advisories/openclaw-before-2026.8…
https://www.vulncheck.com/advisories/openclaw-before-2026.8.1-sandbox-policy-by…