lead-quality-pattern-analysis
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The README.md file includes a command to download the skill's SKILL.md file from the author's repository on GitHub (raw.githubusercontent.com/fourteenwm/...). This is a standard installation process from a well-known service and the vendor's own account.
- [INDIRECT_PROMPT_INJECTION]: The skill processes behavioral data from GA4, such as landing page paths and user metrics, which can contain user-controlled strings. The analysis framework uses this data for classification purposes. Evidence chain:
- Ingestion points: GA4 behavioral data including landing pages, geographic data, and user engagement metrics (SKILL.md).
- Boundary markers: None explicitly defined for the external data.
- Capability inventory: The skill is restricted to the
Readtool for data analysis (SKILL.md frontmatter). - Sanitization: Not explicitly implemented within the skill instructions.
- [SAFE]: The skill consists of domain-specific diagnostic instructions and frameworks. It does not perform network operations, access sensitive files, or execute dynamic code. No malicious patterns or obfuscation techniques were identified.
Audit Metadata