broadcast-campaign

Pass

Audited by Gen Agent Trust Hub on May 16, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill demonstrates the use of template variables (e.g., {{name}}, {{order_id}}) to dynamically insert untrusted external data into message bodies, creating a surface for indirect prompt injection.
  • Ingestion points: Recipient data and template variables are ingested via the contacts array in the zavu.broadcasts.contacts.add method (SKILL.md).
  • Boundary markers: Absent. The provided code examples do not show delimiters or explicit instructions for the model to ignore potential instructions embedded within the template variables.
  • Capability inventory: The skill enables the agent to create, update, and broadcast messages to large numbers of recipients across SMS, Email, WhatsApp, Telegram, Instagram, and Voice (SKILL.md).
  • Sanitization: No explicit sanitization or escaping of the template variables is shown in the snippets, although the documentation notes that message content undergoes an 'AI content review' before transmission.
Audit Metadata
Risk Level
SAFE
Analyzed
May 16, 2026, 07:57 PM
Security Audit — agent-trust-hub — broadcast-campaign