self-improver
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill operates as a purely instructional framework for analyzing affiliate marketing data. It does not request or utilize any sensitive tools, perform network operations, or access restricted file paths.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input, specifically campaign descriptions and qualitative feedback in SKILL.md. Evidence Chain: 1. Ingestion points: campaign.description and results.feedback in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: No tools, subprocess calls, or network operations. 4. Sanitization: Absent.
Audit Metadata