roi-calculator
Pass
Audited by Gen Agent Trust Hub on Aug 20, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill performs expected business logic for ROI calculation using provided templates and standard formulas. It does not request elevated privileges, execute external code, or exfiltrate data.
- [PROMPT_INJECTION]: The skill exhibits a potential surface for indirect prompt injection as it ingests untrusted campaign and results data for report generation. However, the use of structured markdown templates for all outputs significantly mitigates the risk of instructions in the data being misinterpreted as agent commands.
- Ingestion points: Processes campaign spend tables and results data (reach, impressions, revenue) via user input as described in SKILL.md.
- Boundary markers: The skill uses predefined markdown templates in references/roi-templates.md to structure all computed outputs.
- Capability inventory: The skill is limited to writing report files to the memory/ directory.
- Sanitization: No specific input sanitization is performed, but the rigid template structure acts as a logical boundary.
Audit Metadata