extension-invite-links

Pass

Audited by Gen Agent Trust Hub on Nov 2, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill implements a guest RSVP system where unauthenticated users can submit data that is later viewed by administrators.
  • Ingestion points: The submitRSVP function in src/components/GuestRSVP.tsx allows users to provide a name and inviteCode directly via input fields or URL parameters.
  • Boundary markers: There are no delimiters or instructions to ignore embedded commands when the data is stored or retrieved.
  • Capability inventory: The getAllRSVPs function in src/components/AdminDashboard.tsx retrieves and renders all guest-submitted data. An agent using this skill to manage events would read this untrusted content.
  • Sanitization: No input validation or output encoding is demonstrated in the provided frontend snippets for the name field.
  • [EXTERNAL_DOWNLOADS]: The skill relies on external Motoko packages managed via mops.
  • Evidence: The skill requires caffeineai-invite-links and caffeineai-authorization (version ~1.0.0). These are vendor-specific resources associated with the skill's author.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 2, 2026, 09:03 PM