The vulnerability exists in the WebSocket endpoint of gpt-researcher v0.14.7 and earlier. An attacker can send crafted Model Context Protocol messages that lead to arbitrary code execution on the server. No authentication is required, meaning any remote host can trigger the flaw. The impact is complete compromise of the target system, giving the attacker the same privileges as the gpt-researcher process.

Any deployment of gpt-researcher running version 0.14.7 or earlier is affected. This includes installations that expose the WebSocket endpoint publicly or within an organizational network, as no vendor or CPE information distinguishes separate configurations.

The CVSS score is 9.8, indicating a very high severity. EPSS data shows a probability of less than 1%, and the vulnerability is not listed in CISA KEV. The attack vector is a remote WebSocket connection, and no authentication is required, so typical attackers can discover and exploit it quickly. The risk is therefore critical, with a potential for full system compromise.