brand-identity-generator
Pass
Audited by Gen Agent Trust Hub on Mar 26, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill facilitates an indirect prompt injection surface by ingesting untrusted user input and persisting it in the
BRAND-IDENTITY.mdfile, which is explicitly intended to guide the behavior of downstream AI agents (e.g., Claude Code, Cursor, Replit). - Ingestion points: User responses to the 'What are you building?' discovery question and pasted CSS/Tailwind configurations in the 'Expert Path'.
- Boundary markers: Absent. The output template does not include delimiters or instructions for downstream agents to ignore potentially malicious instructions embedded in the user-provided data.
- Capability inventory: The agent uses file-writing capabilities to create and update the
BRAND-IDENTITY.mdfile in the project root. - Sanitization: No sanitization, filtering, or validation of user-provided content is performed before the file is generated.
Audit Metadata