Voice phishing simulations that branch based on what the target actually says

Voice phishing rarely follows a single linear script, a real vishing call adapts based on how the target responds, and Claro's script builder is designed the same way. Scripts are built from typed nodes: an opener sets up the pretext, branch nodes route the call based on the target's DTMF keypress, data-request nodes prompt for information, and outcome nodes close the call. TwiML is generated per node, so a caller's keypress resolves directly to the next node in the script through the same branching logic an experienced social engineer would use on a live call.

Campaigns run in either automated or guided mode, using per-tenant Twilio credentials that are AES-256-GCM encrypted at rest, drawing from a caller-ID pool and respecting a configured call window and timezone so calls only go out at hours your organization has approved. Every call resolves to a specific, meaningful outcome, complied, refused, reported, no-answer, voicemail, partial compliance, or hung up immediately, and that outcome feeds directly into risk scoring, just-in-time training assignment, and gamification, the same behavioral pipeline used for email and WhatsApp simulations. A preflight check validates the script, disclosure language, provider configuration, target coverage, and call window before a campaign can start, and campaigns support pause, resume, and cancel at any point.