churn-prevention
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill includes instructions to read business context from an external file named
marketing-context.md. While this creates an ingestion point for external data, the risk is negligible as the skill lacks high-privilege tools or capabilities that could be exploited by an injected prompt. - Ingestion points: Reads
marketing-context.mdto gather business data (SKILL.md). - Boundary markers: None explicitly defined to separate external data from system instructions.
- Capability inventory: Limited to text generation and running a localized mathematical script.
- Sanitization: None performed on the input file, but output is restricted to safe calculations.
- [COMMAND_EXECUTION]: The skill includes a Python script (
scripts/churn_impact_calculator.py) designed for revenue modeling. Technical analysis of this script confirms it is restricted to mathematical processing and does not perform any dangerous operations, such as network calls, arbitrary code execution, or access to sensitive system files.
Audit Metadata