agency-behavioral-nudge-engine
Pass
Audited by Gen Agent Trust Hub on Jul 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a specialized persona and instruction set for behavioral coaching. All analyzed content adheres to standard agent configuration patterns.
- [DATA_EXPOSURE]: The skill mentions memory and tracking of user preferences such as communication channels (SMS/Email) and motivational triggers. These are functional requirements for the stated purpose and do not involve unauthorized access to sensitive system files or credentials.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data (task lists and user profiles) to generate nudges. While this represents a potential ingestion point for untrusted data, the skill's capabilities are limited to generating text prompts and reinforcement copy, presenting a minimal attack surface. No dangerous tools or command execution capabilities are requested.
- [METADATA_POISONING]: The metadata in the YAML configuration accurately reflects the skill's purpose, including localized display names that match the English description.
Audit Metadata