ai-graphic-design
Pass
Audited by Gen Agent Trust Hub on Apr 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill contains instructional content and workflow documentation for graphic design tasks. No malicious code, obfuscation, or unauthorized data access patterns were detected.
- [EXTERNAL_DOWNLOADS]: The documentation references several well-known AI design platforms and tools (e.g., Midjourney, Adobe Firefly, Recraft, Vectorizer.ai) and identifies Python libraries for automation. These references are appropriate for the skill's stated purpose and target established professional services.
- [COMMAND_EXECUTION]: Section 9 describes a Python-based automation pipeline using libraries like
py5andvpypefor batch processing SVG files. These are standard operations for design automation and do not involve unauthorized or hidden execution. - [INDIRECT_PROMPT_INJECTION]: The skill establishes a data ingestion surface by instructing the agent to process user-provided brand manuals and strategy documents (e.g., RCAO and StoryBrand frameworks). While this involves processing untrusted data, it is a core feature of the design workflow and does not grant the skill excessive capabilities.
Audit Metadata