DEV Community

CVE Reports
CVE Reports

Posted on • Originally published at cvereports.com

GHSA-M8V2-6WWH-R4GC: GHSA-M8V2-6WWH-R4GC: Sandbox Escape via Symlink Manipulation in OpenClaw

GHSA-M8V2-6WWH-R4GC: Sandbox Escape via Symlink Manipulation in OpenClaw

Vulnerability ID: GHSA-M8V2-6WWH-R4GC
CVSS Score: 8.2
Published: 2026-03-03

A critical logic error in OpenClaw's path validation mechanism allows attackers to bypass sandbox restrictions. By requesting bind mounts for non-existent files within symlinked parent directories, attackers can trick the validation logic into accepting paths that resolve to restricted host directories (e.g., /etc or /var). This effectively breaks the security boundary between the sandboxed agent and the host system.

TL;DR

OpenClaw failed to resolve symlinks for paths where the final file (leaf) did not exist. Attackers can exploit this by creating symlinks to sensitive host directories and mounting non-existent files through them, achieving arbitrary file write access on the host.


⚠️ Exploit Status: POC

Technical Details

  • CWE ID: CWE-59 (Link Following)
  • Attack Vector: Local (Sandbox Escape)
  • Impact: Host Filesystem Read/Write
  • CVSS Score: 8.2 (High)
  • Exploit Status: PoC Available
  • Affected Component: validateBindMounts

Affected Systems

  • OpenClaw AI Agent Platform
  • OpenClaw: <= 2026.2.23 (Fixed in: Post-2026.2.23)

Code Analysis

Commit: b5787e4

Fix: correctly resolve parent paths for non-existent files to prevent sandbox escape

Mitigation Strategies

  • Update OpenClaw to the latest stable release post-2026.2.23.
  • Restrict agent permissions to prevent symlink creation if not required.
  • Audit allowedSourceRoots configuration.

Remediation Steps:

  1. Stop the OpenClaw service.
  2. Pull the latest code or docker image containing commit b5787e4abba0dcc6baf09051099f6773c1679ec1.
  3. Verify the version is > 2026.2.23.
  4. Restart the service.

References


Read the full report for GHSA-M8V2-6WWH-R4GC on our website for more details including interactive diagrams and full exploit analysis.

Top comments (0)