ab-testing
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill contains no executable scripts, commands, or sensitive file access. It is purely instructional, providing templates and frameworks for designing and analyzing experiments.
- [EXTERNAL_DOWNLOADS]: The skill includes links to well-known and trusted experimentation tools (Evan Miller, Optimizely, VWO, and AB Test Guide). These are provided as resources for the user to perform manual calculations and do not involve automated code downloads or script execution.
- [INDIRECT_PROMPT_INJECTION]: The skill instructions suggest reading local project context files (e.g.,
.agents/product-marketing.md) to inform test design. - Ingestion points: Reads from
.agents/product-marketing.md,.claude/product-marketing.md, orproduct-marketing-context.mdas specified inSKILL.md. - Boundary markers: None identified.
- Capability inventory: No risky capabilities (such as shell execution, network operations, or file-write tools) are used by this skill.
- Sanitization: None identified. Given the lack of dangerous tools, this ingestion point does not pose a security risk.
Audit Metadata