value-proposition
Pass
Audited by Gen Agent Trust Hub on Jul 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely composed of natural language instructions and does not contain any executable scripts, code blocks, or automated command patterns.
- [DATA_EXFILTRATION]: The skill instructs the agent to read from
.agents/product-marketing.md. This is a local configuration file used to provide context for the marketing task and does not involve unauthorized access to system-level sensitive data or credentials. - [PROMPT_INJECTION]: The skill processes data from a local file and user input to generate its outputs. While this represents a surface for indirect prompt injection (Category 8), the risk is mitigated by the skill's lack of tools or capabilities that would allow for network exfiltration or command execution. It follows standard agent workflow practices for data ingestion.
Audit Metadata