software-mobile
Pass
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill contains prescriptive instructions within the "Trend Awareness Protocol" section that mandate specific agent behaviors and web search triggers based on user input topics.
- [PROMPT_INJECTION]: The skill identifies an attack surface for indirect prompt injection by utilizing user-provided input to drive external search tool actions.
- Ingestion points: User queries regarding framework comparisons, trends, and platform recommendations in SKILL.md.
- Boundary markers: None provided to separate user-provided queries from the agent's internal instructions.
- Capability inventory: Uses web search capabilities to fetch and process external data.
- Sanitization: No mechanisms for sanitizing or validating user input are specified before the input is used to trigger search actions.
Audit Metadata