The flaw in OpenClaw’s host execution environment filtering can allow a caller to set interpreter startup variables without proper authorization checks. When the vulnerable feature is enabled, a low‑trust caller or an attacker controlling an input path can insert crafted environment variables, enabling execution or persistence of actions beyond the intended authorization, effectively bypassing authentication controls and enabling unauthorized code execution.

OpenClaw versions earlier than 2026.6.6 are affected. The vulnerability applies to the OpenClaw application running on Node.js environments. The known CNA vendor product is OpenClaw:OpenClaw, with the affected range including all releases prior to 2026.6.6.

The CVSS score of 8.7 indicates high severity. EPSS score of < 1% suggests a very low likelihood of exploitation, and the vulnerability is not currently listed in the CISA KEV catalog. Based on the description, the likely attack vector involves a low‑trust caller supplying crafted environment variables to the relevant feature; this could be achieved remotely if the feature is exposed over a network interface, or locally if a local attacker can influence the input path. Successful exploitation would grant the attacker bypass of authentication controls and execution privileges that exceed the original caller’s authority. Since no mitigation is in place by default, an attacker with access to set environment variables could easily achieve the exploit.