eighty-twenty-customer-value-rfm
Pass
Audited by Gen Agent Trust Hub on Jun 21, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is entirely composed of markdown instructions and theoretical frameworks for marketing analysis. It contains no executable code, network-enabled tools, or file system operations.
- [PROMPT_INJECTION]: The skill defines a workflow that involves ingesting customer and behavioral data (Step 1: Define The Customer Set). While this creates a potential surface for indirect prompt injection if an agent processes untrusted data, the skill does not define any capabilities to exploit such an injection.
- Ingestion points: User-provided customer lists and behavioral metrics described in SKILL.md.
- Boundary markers: No specific delimiters or safety instructions are included to isolate ingested data from agent instructions.
- Capability inventory: The skill defines no tool usage, shell command execution, or network requests.
- Sanitization: No validation or sanitization steps are provided for the input data.
Audit Metadata