The vulnerability allows an authenticated user to bypass the commands.allowFrom verification in Telegram interactive callbacks, enabling them to execute commands outside of the intended sender restrictions. This can lead to unintended command execution and privilege escalation within the bot’s environment. The weakness is identified as CWE‑863.

OpenClaw versions released before 2026.5.6, run on platforms supported by the product, are affected. The issue resides in the Telegram integration component that processes interactive callbacks.

The CVSS score of 7.7 signifies a high severity security flaw. Because the EPSS score is unavailable, the probability of real‑world exploitation cannot be determined, but the vulnerability is not listed in CISA's KEV catalog. Attackers require valid authentication to the bot and can trigger the bypass by sending a crafted callback message, then setting themselves as an authorized sender through the callback before global allowlist checks are applied. The risk is that bot administrators could be subverted to run arbitrary commands.