positioning-statement
Pass
Audited by Gen Agent Trust Hub on Jul 29, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests and processes untrusted data from both user inputs and local environment files.
- Ingestion points: The skill reads user-supplied product details (target, problem, benefit) and instructions specifically suggest reading project context files such as
.agents/product-marketing.mdand.agents/aaj-brand.md. - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are used when interpolating external data into the prompts.
- Capability inventory: The skill contains no executable code, scripts, or tool definitions; its capabilities are limited to text generation within the agent's context.
- Sanitization: There is no logic provided to sanitize, validate, or filter user-provided content before it is processed.
Audit Metadata