PraisonAI is a multi-agent teams system. Prior to praisonaiagents 1.6.58, the FileMemory constructor joins unsanitized user_id into self.user_path. A caller supplying ../ or path separators can escape the memory directory and write JSON data to arbitrary process-writable locations. The fix sanitizes user_id before constructing self.user_path. This issue is fixed in version 1.6.58.

CVE-2026-55527 affects PraisonAI, a multi-agent teams system, in versions prior to praisonaiagents 1.6.58. The vulnerability exists in the FileMemory constructor, which joins an unsanitized user_id parameter into self.user_path without proper validation. An attacker supplying path traversal sequences such as '../' or other path separators can escape the intended memory directory boundary. This allows writing arbitrary JSON data to any location writable by the process, potentially enabling unauthorized file writes across the filesystem. The vulnerability is classified as a path traversal (directory traversal) issue. The fix was implemented by sanitizing the user_id input before it is used to construct self.user_path. The issue has been patched in version 1.6.58 of praisonaiagents, with the fix committed to the official GitHub repository.