Disclosed weakness · no CVE recorded
OpenClaw could lose read-only protection for nested skill directories
OpenClaw · OpenClaw · sandbox skill storage
First documented disclosure · Review checkpoint
Summary
A sandbox with a writable workspace could replace content at a protected nested skill location.
Severity and exploitation evidence
No CVSS score recorded in this review.
Exploitation not established: The upstream sources document the behavior and correction. They identify no malicious use.
Technical severity is not a measure of observed attacks or of risk to your deployment. Attempts, demonstrations, and successful compromise retain the qualifications in the source.
How the weakness works
The protection covered the skill location but left its surrounding directory structure mutable.
Conditions required
A sandboxed task can write to a workspace containing nested protected skill directories.
Potential impact
Later reads could consume content different from the operator-managed skill.
Affected versions and configurations
The upstream report identifies the affected behavior without specifying a complete version range.
Fix and mitigation status
The 2026.11.4 stable release includes this issue-specific correction.
Practical next steps
Deploy the fix and compare managed skill contents with trusted copies before reuse.
Operating lesson
Test the integrity of a protected file across its full directory hierarchy.
Evidence and limits
This concerns already-mounted nested paths. Missing directories on first launch are tracked separately.
Sources
- OpenClaw · fix(sandbox): pin writable ancestors of nested protected skill mounts
Primary source ·
- OpenClaw 2026.11.4 stable release notes
Primary source ·