human-taste-content
Pass
Audited by Gen Agent Trust Hub on Feb 27, 2026
Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists of instructional Markdown files that provide a rubric for content quality assessment. No malicious behaviors, obfuscation techniques, or unauthorized data access attempts were identified during the analysis.
- [NO_CODE]: No executable scripts, binaries, or configuration files are present in the skill package; it is purely instructional and informational.
- [PROMPT_INJECTION]: The skill is designed to evaluate untrusted text provided by the user, which creates a surface for indirect prompt injection. However, since the skill possesses no sensitive capabilities such as network communication, file system access, or command execution, the potential impact is negligible and limited to influencing the AI's textual critique. \n
- Ingestion points: Processes user-provided copy, documentation, and marketing text for evaluation as described in SKILL.md. \n
- Boundary markers: No specific delimiters or isolation techniques are instructed for the input content. \n
- Capability inventory: No scripts, subprocess calls, or network operations are included in the skill files. \n
- Sanitization: No input sanitization or validation of the provided text is performed.
Audit Metadata