brand-voice-generator
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the user (such as brand descriptions, design principles, and voice attributes) and interpolates it directly into configuration and documentation files.
- Ingestion points: Interactive user responses during the brand setup process (SKILL.md Steps 1 through 6).
- Boundary markers: The templates in
references/brand-template.mdandreferences/tone-template.mdlack boundary markers or explicit instructions to ignore embedded commands within the generated content. - Capability inventory: The skill possesses file-writing capabilities to the
.claude/skills/pptx-generator/brands/directory. - Sanitization: No input validation or sanitization is performed on the user-provided strings before they are written to the JSON and Markdown files. This could allow a malicious user to embed instructions that influence the behavior of downstream 'brand-aware' skills that process these files.
Audit Metadata