design-prompt-builder
Pass
Audited by Gen Agent Trust Hub on Aug 19, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill constructs prompts for a downstream AI session by interpolating untrusted user input into structured templates without robust sanitization.
- Ingestion points: The workflow in
SKILL.md(Workflow Step 3) captures natural language input from the user regarding product descriptions, goals, and constraints to populate the prompt template. - Boundary markers: The final prompt synthesis (Workflow Step 4) uses simple headers like
[CONTEXT]but lacks strong encapsulation (such as XML tags or triple delimiters) to isolate user input from the rest of the prompt instructions. - Capability inventory: The skill itself is limited to text generation and does not possess capabilities for file modification, network operations, or command execution.
- Sanitization: There are no visible filters or safety instructions to prevent the agent from obeying malicious instructions embedded within the user-provided context fields.
Audit Metadata