churn-breakdown
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external billing reports which include user-controlled text fields like cancellation reasons and plan names.
- Ingestion points: Untrusted data enters the context via ZohoBilling_Get_Voluntary_Churn_Details_Report and ZohoBilling_Get_Involuntary_Churn_Details_Report as defined in SKILL.md.
- Boundary markers: The instructions lack explicit delimiters or instructions to the agent to disregard potentially malicious instructions embedded in the report data.
- Capability inventory: The skill possesses no dangerous capabilities; all scripts and instructions are limited to read-only data fetching and summary generation, with no tools for file modification, network communication, or shell access.
- Sanitization: No sanitization or validation steps are defined for the report content prior to processing.
Audit Metadata