hyva-ui-component

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local shell commands (including ls, find, and cp) to manage component files and theme structures. It also invokes a local helper script, scripts/refresh_catalog.sh, to regenerate the component catalog based on the installed vendor package.
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for installing official vendor packages (hyva-themes/hyva-ui, hyva-themes/magento2-hyva-payment-icons) using Composer. It also references the official vendor download portal at hyva.io.
  • [INDIRECT_PROMPT_INJECTION]: The skill reads and processes README.md files from the hyva-ui component package to display instructions and extract XML configuration.
  • Ingestion points: Component README.md files located within the vendor/hyva-themes/hyva-ui directory.
  • Boundary markers: No explicit delimiters or boundary markers are defined in the instructions for separating external content from agent instructions.
  • Capability inventory: The skill has capabilities for file system writes (cp), shell command execution, and local script execution.
  • Sanitization: The skill extracts specific XML blocks based on patterns but does not sanitize the descriptive markdown content before it is processed or presented to the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 06:41 AM
Security Audit — agent-trust-hub — hyva-ui-component