AI security incident: OpenClaw exec approvals: safeBins could bypass stdin-only constraints via shell expan...
Summary OpenClaw's exec-approvals allowlist supports a small set of "safe bins" intended to be stdin-only (no positional file arguments) when running tools.exec.host=gateway|node with security=allowlist . In affected configurations, the allowlist validation checked pre-expansion argv tokens, but execution used a real shell ( sh -c ) which expands globs and environment variables. This allowed safe bins like head , tail , or grep to read arbitrary local files via tokens such as * or $HOME/... without triggering approvals. This issue is configuration-dependent and is not exercised by default settings (default tools.exec.host is sandbox ). ## Affected Packages / Versions - Package: openclaw (npm) - Affected: = 2026.2.
Why This Is AI-Related
This advisory is part of the public incident archive, but the current source material uses limited explicit AI terminology, so the cited sources should be reviewed carefully when judging AI relevance and exposure.
- Explicit AI-specific signals are limited in the current source material, so use the cited advisory to validate scope during triage.
Affected Workflow
Review the AI product, dependency, and integration points mentioned in the source advisory before broadening remediation.
Likely Attack Path
The advisory indicates a security path that can affect AI applications, assistants, models, or connected automation workflows if the component is deployed.
Impact
The advisory has meaningful security implications for an AI-related product, dependency, or workflow and should be triaged against deployed usage. Severity HIGH. Classification confidence 50%. Source channel GHSA.
Detection And Triage Signals
- New security events tied to the affected component or advisory identifier
- Changes in AI workflow behavior, access logs, or plugin execution after the advisory window
- Evidence that the vulnerable version is active in environments that process sensitive data
Recommended Response
- Confirm whether affected products, models, or integrations are used in your environment.
- Apply vendor fixes or mitigations and restrict risky permissions until verified.
- Monitor logs for related indicators and document containment for audit evidence.
Compliance And Business Impact
Even when exploit details are still emerging, delayed triage can widen operational and compliance exposure around AI systems.
Sources
Want alerts like this in real time?
Get notified with incident context, likely impact, and response guidance.
Get Notified