Impact
The vulnerability in the ClawHub application lies in its public profile image preview feature, which accepts a user‑supplied URL and only checks the textual hostname against private‑address patterns. It does not verify the actual resolved IP address, allowing an attacker to provide a hostname that resolves to an internal network address. This produces a server‑side request forgery that can establish outbound connections to internal services. Although the maintainer ran a proof‑of‑concept that created an outbound connection to a loopback listener, no internal service access, credential disclosure, or code execution was demonstrated; nevertheless the potential for those outcomes exists.
Affected Systems
OpenClaw’s ClawHub product is affected. Any self‑hosted deployment running a version prior to the fix in commit 8c2de6c506bb4efabe3f0c2ffb8370b9e23d4650 is vulnerable. The exact historical affected range is not fully defined, but the vulnerability was confirmed at revision cbfee7343ddc867316dd9b3de6fa8856730f9f41 and fixed by PR #3683. The npm CLI and OpenClaw runtime are not impacted.
Risk and Exploitability
The CVSS score of 6.9 indicates a moderate risk. The EPSS score is not available, and the vulnerability is not listed in CISA’s KEV catalog. Based on the description, it is inferred that no authentication is required to trigger the flaw via a malicious URL in a public profile preview. A remote attacker can target internal resources, but the exploit would be limited to outbound connections initiated by ClawHub.
OpenCVE Enrichment