CVE-2026-57122

PraisonAI, an open-source multi-agent teams framework, fails to enforce HMAC signature verification on its WhatsApp and Linear bot webhook handlers in versions prior to 4.6.59. When WHATSAPP_APP_SECRET or LINEAR_WEBHOOK_SECRET is not configured, the handlers accept, parse, and dispatch completely unsigned request bodies. Any remote unauthenticated attacker who can reach the webhook route can forge incoming messages, comments, or agent-session events, impersonate legitimate platform users, and steer agent prompts and actions, as well as disrupt bot processing. Organizations running self-hosted PraisonAI deployments with these bot integrations exposed are affected; the issue is fixed in version 4.6.59. No public proof of concept or known in-the-wild exploitation has been reported, and the CVE is not in the CISA KEV catalog.

What to do: Upgrade PraisonAI to 4.6.59 or later immediately. Ensure WHATSAPP_APP_SECRET and LINEAR_WEBHOOK_SECRET are set on every deployment so HMAC verification is actually enforced, and confirm webhook routes are not reachable from untrusted networks (restrict via reverse-proxy allowlists of WhatsApp/Linear source ranges). Review bot and agent session logs for forged messages, comments, or anomalous session events that may indicate prior abuse.