cro
Pass
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues were detected. The skill consists of text-based instructions for providing conversion rate optimization advice based on marketing best practices.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow for analyzing external web pages and URLs provided by users, which creates a surface for indirect prompt injection. This is an inherent property of a conversion analysis tool. The risk is mitigated because the skill lacks the ability to execute system commands, write to the filesystem, or perform unauthorized network operations. 1. Ingestion points: User-provided URLs and page content mentioned in SKILL.md and evals.json. 2. Boundary markers: Absent. No specific instructions are provided to the agent to isolate or ignore instructions embedded within the analyzed marketing content. 3. Capability inventory: Text-based analysis and recommendation generation. No dangerous tools are requested or used. 4. Sanitization: Absent. Page content is processed directly for analysis.
Audit Metadata