skills/naimsolong/ai-software-agency/specialist-marketing-cross-border-ecommerce/Gen Agent Trust Hub
specialist-marketing-cross-border-ecommerce
Pass
Audited by Gen Agent Trust Hub on Aug 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a prompt for a marketing sub-agent using the
Agenttool. No malicious code, obfuscation, or suspicious patterns were detected in the instructions or metadata. - [PROMPT_INJECTION]: The skill accepts user input through the
{the user's question}placeholder, which is a standard interface for agent-based skills. The absence of high-risk capabilities like file system access, network operations, or arbitrary command execution minimizes the risk associated with this injection surface. - Ingestion points: The user input is interpolated into the
Agenttool prompt inSKILL.mdvia{the user's question}. - Boundary markers: None are explicitly defined around the user input placeholder.
- Capability inventory: The skill is restricted to using the
Agenttool to spawn a sub-agent. No subprocess, file system, or network capabilities are present. - Sanitization: No input sanitization or filtering logic is specified in the skill definition.
Audit Metadata