gc-minimal-zine-poster-v0-3
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill package is composed entirely of markdown instructions and configuration files. It does not include any scripts, binaries, or external dependencies, relying on the host platform's native tools for image generation and file inspection.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user content, such as articles and reference images, creating a potential surface for indirect injection. The risk is mitigated by specific instructions: 1. Ingestion points: User-supplied themes, articles, and image folders processed in SKILL.md and references/reference-analysis.md. 2. Boundary markers: Requirements to isolate and exclude 'sample residue' like source brands and watermarks. 3. Capability inventory: Uses the host's built-in image tools. 4. Sanitization: Steps to derive metaphors and visual rules while ignoring literal source text.
Audit Metadata