bx-ui-forms
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where untrusted data from external sources is interpolated directly into HTML form components.
- Ingestion points: External data from the
urlandformscopes (such as#url.maxPrice#and#form.fullName#) is used within component attributes inSKILL.md. - Boundary markers: The documentation lacks explicit instructions or delimiters to ensure the agent treats this interpolated content as potentially malicious data.
- Capability inventory: The skill utilizes form components (
bx:input,bx:slider) to render this data into the user's browser context. - Sanitization: There is no explicit documentation regarding whether the
bx-ui-formscomponents or the BoxLang engine perform automatic HTML escaping or sanitization to prevent XSS attacks. - [COMMAND_EXECUTION]: The skill instructions include shell commands for environment setup and module installation.
- Evidence: Commands
install-bx-module bx-ui-formsandbox install bx-ui-formsare listed in the Installation section ofSKILL.md. - Context: These are standard administrative actions for managing modules within the BoxLang and CommandBox ecosystems.
Audit Metadata