email-finding
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion and processing of untrusted contact data, creating an attack surface for indirect prompt injection.
- Ingestion points: Contact data (names, company domains, LinkedIn URLs) enters the agent's context through user request prompts or file-based batch enrichment as described in Phase 1 (Intake) of
SKILL.md. - Boundary markers: The skill lacks explicit instructions for using delimiters or boundary markers to isolate untrusted contact data from the agent's primary instructions.
- Capability inventory: The skill enables interactions with multiple external B2B data providers (LeadMagic, Apollo, ZoomInfo, People Data Labs, Hunter.io, and Claygent) to perform data enrichment and lookups.
- Sanitization: There are no specified sanitization or validation steps to filter or escape potentially malicious payloads embedded within the contact data before it is utilized in lookups or AI-assisted discovery processes.
Audit Metadata