sent-contacts
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes contact information retrieved from external sources, which represents a potential attack surface for indirect prompt injection. However, the skill implements robust mitigations through mandatory human-in-the-loop confirmation gates for all sensitive operations.
- Ingestion points: Contact records fetched via contacts.list and contacts.get in SKILL.md.
- Boundary markers: No explicit prompt delimiters are defined, but instructions mandate local data validation.
- Capability inventory: Includes powerful mutation tools such as contacts.create_many and contacts.delete.
- Sanitization: The skill requires clear payload previews and explicit user confirmation immediately before any data mutation, preventing unauthorized or automated execution of embedded instructions.
Audit Metadata