imessage-management

Pass

Audited by Gen Agent Trust Hub on Sep 21, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input and external tool outputs while maintaining capabilities for financial transactions and outbound messaging.
  • Ingestion points: The skill ingests user-supplied strings for first/last names and public image URLs for profile photos via set_imessage_display_name and set_imessage_profile_photo. It also processes price and eligibility data from get_imessage_offer.
  • Boundary markers: There are no explicit delimiters or instructions to ignore potential instructions embedded within user-supplied identity fields.
  • Capability inventory: The skill possesses the ability to provision numbers that incur recurring costs (provision_imessage_number), send outbound iMessages to leads (send_message), and modify public-facing profile metadata.
  • Sanitization: A critical manual safety check is implemented requiring the agent to display the confirmed price and obtain explicit human confirmation before the accept_charge: true parameter is used in provisioning tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 21, 2026, 01:29 PM