CVE-2026-61428 describes an authentication bypass vulnerability in MervinPraison's PraisonAI AgentMail product. Versions prior to 4.6.78 do not verify signatures on incoming webhook messages, allowing unauthenticated attackers to spoof sender addresses by posting crafted message.received events. This enables injection of arbitrary content into the agent and causes it to send replies to attacker-controlled addresses, circumventing sender allow/block lists. The CVSS 4.0 base score is 6.9 (medium severity), reflecting network attack vector, no privileges or user interaction required, and low to limited impact on confidentiality, integrity, and availability.

An attacker can inject arbitrary messages into the PraisonAI AgentMail system by spoofing sender addresses without authentication. This can lead to the system sending replies to attacker-controlled addresses, bypassing sender allow/block lists. The vulnerability impacts message integrity and could facilitate phishing or spam campaigns using the compromised agent. There is no indication of direct system compromise or data exfiltration beyond message injection.

Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. No official fix or patch information is provided. Until a patch is available, organizations should consider restricting access to the webhook endpoint or implementing additional verification controls externally to prevent unauthenticated message injection.