orderly-one-dex
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines parameters for DEX customization, including
analyticsScriptandthemeCSS, which accept user-supplied content. While this creates a potential surface for injecting malicious code into the generated DEX frontend, this is an inherent feature of the white-labeling platform described. - Ingestion points:
analyticsScript,themeCSS,customMenus, and branding file uploads inSKILL.md. - Boundary markers: None specified in the API documentation.
- Capability inventory: The skill enables the agent to interact with the Orderly One REST API for DEX deployment and management.
- Sanitization: The documentation does not specify server-side sanitization, placing the responsibility on the agent or user to provide valid, non-malicious content.
- [DYNAMIC_EXECUTION]: The platform supports the inclusion of Base64-encoded
analyticsScriptdata which is executed on the deployed DEX frontend. This is a functional requirement for the service and does not represent a risk to the agent's local environment.
Audit Metadata