nexudus
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external API responses (Nexudus) that may contain user-generated content such as delivery notes, visitor names, or event descriptions.
- Ingestion points: Data is fetched via
httpxfromhttps://*.spaces.nexudus.com/api/public/endpoints as defined insrc/nexudus/__init__.py. - Boundary markers: The skill does not use specific delimiters or instructions to treat the API output as untrusted data when passing summarized or raw records to the agent.
- Capability inventory: The skill includes functions for state-changing operations such as
mark_collected,invite_visitor,create_booking, andcancel_booking(POST, PUT, and DELETE methods). - Sanitization: Output is truncated for length via the
_clipfunction insrc/nexudus/__init__.py, but no content sanitization, escaping, or filtering of potential prompt injection patterns is performed.
Audit Metadata