Banks generates meaningful LLM prompts using a simple template language. In versions prior to 2.4.4, all four media filters (image, audio, video, document) in b…

CVE-2026-62663 is a path traversal vulnerability in the 'banks' Python library, which generates LLM prompts using a template language. In versions prior to 2.4.4, all four media filters (image, audio, video, document) accept untrusted user input as file paths and pass them directly to open() without any sanitization or directory restrictions. An attacker who controls template variables can use ../ sequences to read arbitrary files accessible to the Python process, including .env files, SSH keys, cloud credentials, /etc/passwd, and /etc/shadow. The exfiltrated file contents are returned base64-encoded in the rendered prompt output, making data exfiltration trivial. This is especially dangerous in applications that process user-provided template variables before sending prompts to an LLM. The vulnerability has been patched in version 2.4.4.