TypeBot versions before 3.16.0 contain an SSRF vulnerability in the OpenAI "Create Transcription" action handler. The handler uses fetch() on a user-supplied audio URL without the SSRF protections that are present elsewhere in the codebase. This flaw enables attackers to coerce the server into making HTTP requests to internal network addresses or localhost. The server then forwards the fetched audio content to the OpenAI Whisper API for transcription, and the transcription output is returned to the attacker. This can lead to unauthorized internal network probing and data exposure. The vulnerability is fixed in version 3.16.0.

An attacker can exploit this SSRF vulnerability to make the server send HTTP requests to arbitrary internal or localhost addresses. This could allow internal network reconnaissance or access to internal services not normally exposed externally. The attacker also receives the transcription result from the OpenAI Whisper API based on the fetched content, potentially leaking sensitive information. The CVSS score of 5.4 (medium severity) reflects limited confidentiality and integrity impact with no availability impact.

Upgrade TypeBot to version 3.16.0 or later, where this SSRF vulnerability is fixed. No other official remediation or temporary fixes are documented. Until upgraded, avoid using the OpenAI "Create Transcription" action handler with untrusted audio URLs to prevent exploitation.