The Lettre library versions before 0.11.24 have a vulnerability in their STARTTLS implementation that allows a man-in-the-middle attacker to inject forged SMTP replies. This occurs because the attacker can append their own replies to the cleartext STARTTLS response, which the client then mistakenly treats as legitimate server responses after the TLS upgrade. This affects only connections upgraded with STARTTLS, not implicit TLS connections.