brand-visual-generator
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of instructional text guiding the AI agent on how to recommend design systems, typography, and color palettes. It does not include any scripts, commands, or executable code.
- [INDIRECT_PROMPT_INJECTION]: The skill identifies ingestion points for untrusted data, specifically reading project context from
.claude/project-context.mdand.cursor/project-context.md. - Ingestion points: Reads visual identity data from project context files.
- Boundary markers: Absent.
- Capability inventory: None. The skill has no file-write, network exfiltration, or shell execution capabilities.
- Sanitization: Absent.
- Conclusion: Despite the ingestion surface, the lack of dangerous capabilities makes this pattern safe.
- [SAFE]: No obfuscation, persistence mechanisms, or credential exposure patterns were detected in the metadata or instructional body.
Audit Metadata