creator-invoice-readiness
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from external systems, which could contain instructions intended to influence the agent's behavior.
- Ingestion points: As described in
SKILL.md, the agent identifies and checks data from sponsor records, CRM, and banking tools. - Boundary markers: The instructions do not define delimiters to separate system instructions from the untrusted data being processed.
- Capability inventory: The skill instructions in
SKILL.mdindicate the agent may have access to sensitive external tools and the ability to perform actions (like sending invoices or reminders) with user approval. - Sanitization: There are no defined procedures for escaping or validating the content retrieved from external tools before it is processed by the agent.
Audit Metadata