GHSA-grc3-2j34-p6gm: OpenClaw: message.action forwarding could send Gateway credentials to model-supplied loopback URLs
OpenClaw versions prior to 2026.5.2 have a vulnerability where message.action forwarding can send Gateway credentials to model-supplied loopback URLs. This occurs when model-controlled action metadata selects a loopback Gateway URL, causing the action payload and credentials to be forwarded to that URL. The vulnerability is limited to the named feature and configuration and does not affect the overall trusted-operator model. Exploitation depends on the operator's configuration and whether untrusted input can reach the affected path. The issue is patched in version 2026.5.2.
AI Analysis
Technical Summary
In OpenClaw versions before 2026.5.2, the message.action forwarding feature can inadvertently send Gateway credentials and action payloads to loopback URLs supplied by the model. This happens when model-controlled action metadata selects a loopback Gateway URL, forwarding sensitive data to that local listener. The vulnerability is scoped to this feature and does not alter the trusted-operator model involving authenticated Gateway operators and installed plugins. The practical impact depends on the operator's configuration and exposure to lower-trust inputs. The vulnerability is addressed in version 2026.5.2.
Potential Impact
If the affected feature is enabled and reachable, Gateway tokens and action payloads could be exposed to a local listener chosen via model-supplied loopback URLs. This could lead to credential exposure within the local environment. The actual risk depends on the configuration of the Gateway operator and whether untrusted inputs can access the vulnerable forwarding path.
Mitigation Recommendations
A fix is available in OpenClaw version 2026.5.2. Until patched, restrict message.action forwarding and avoid using model-supplied loopback targets. Additionally, narrow channel and tool allowlists, avoid sharing a Gateway between mutually untrusted users, and disable the affected feature when not needed to reduce risk.
GHSA-grc3-2j34-p6gm: OpenClaw: message.action forwarding could send Gateway credentials to model-supplied loopback URLs
Description
OpenClaw versions prior to 2026.5.2 have a vulnerability where message.action forwarding can send Gateway credentials to model-supplied loopback URLs. This occurs when model-controlled action metadata selects a loopback Gateway URL, causing the action payload and credentials to be forwarded to that URL. The vulnerability is limited to the named feature and configuration and does not affect the overall trusted-operator model. Exploitation depends on the operator's configuration and whether untrusted input can reach the affected path. The issue is patched in version 2026.5.2.
CVSS v4.0
Affected software
Run on your own infrastructure? Check whether these packages are installed with threat-finder — our free open-source scanner.
Weaknesses
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
In OpenClaw versions before 2026.5.2, the message.action forwarding feature can inadvertently send Gateway credentials and action payloads to loopback URLs supplied by the model. This happens when model-controlled action metadata selects a loopback Gateway URL, forwarding sensitive data to that local listener. The vulnerability is scoped to this feature and does not alter the trusted-operator model involving authenticated Gateway operators and installed plugins. The practical impact depends on the operator's configuration and exposure to lower-trust inputs. The vulnerability is addressed in version 2026.5.2.
Potential Impact
If the affected feature is enabled and reachable, Gateway tokens and action payloads could be exposed to a local listener chosen via model-supplied loopback URLs. This could lead to credential exposure within the local environment. The actual risk depends on the configuration of the Gateway operator and whether untrusted inputs can access the vulnerable forwarding path.
Mitigation Recommendations
A fix is available in OpenClaw version 2026.5.2. Until patched, restrict message.action forwarding and avoid using model-supplied loopback targets. Additionally, narrow channel and tool allowlists, avoid sharing a Gateway between mutually untrusted users, and disable the affected feature when not needed to reduce risk.
Technical Details
- Gcve Source
- db.gcve.eu
- Osv Id
- GHSA-grc3-2j34-p6gm
- Osv Schema Version
- 1.4.0
- Aliases
- []
- Ecosystems
- ["npm"]
- Database Specific Severity
- MODERATE
- Cvss Version
- 4.0
Threat ID: 6a46ecca27e9c7971943ee53
Added to database: 07/02/2026, 22:57:14 UTC
Last enriched: 07/02/2026, 23:20:18 UTC
Last updated: 07/02/2026, 23:20:18 UTC
Views: 2
Community Reviews
0 reviewsCrowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.
Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.
Actions
Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.
External Links
Need more coverage?
Upgrade to Pro Console for AI refresh and higher limits.
For incident response and remediation, OffSeq services can help resolve threats faster.
Latest Threats
Check if your credentials are on the dark web
Instant breach scanning across billions of leaked records. Free tier available.