heroui-native

Fail

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The SKILL.md file contains an installation command curl -fsSL https://heroui.com/install | bash. This pattern downloads and executes a shell script directly from the vendor's domain. While this is a common installation pattern for developer tools, it executes code outside of a package manager's control.
  • [INDIRECT_PROMPT_INJECTION]: The skill uses scripts to fetch MDX documentation from external vendor endpoints (heroui.com and native-mcp-api.heroui.com), which is then processed by the agent.
  • Ingestion points: Documentation is retrieved in scripts/get_component_docs.mjs and scripts/get_docs.mjs.
  • Boundary markers: No delimiters or warnings are used to wrap the external content.
  • Capability inventory: The agent has access to standard tools for file system and shell operations.
  • Sanitization: The retrieved MDX content is not sanitized for potential instructions before being ingested into the agent's context.
  • [EXTERNAL_DOWNLOADS]: The skill scripts perform network requests to the vendor's API and documentation site to retrieve component information and theme settings.
  • [COMMAND_EXECUTION]: The skill provides Node.js scripts that execute network requests and process command-line arguments to fetch specific documentation components.
Recommendations
  • HIGH: Downloads and executes remote code from: https://heroui.com/install - DO NOT USE without thorough review
Audit Metadata
Risk Level
HIGH
Analyzed
Sep 15, 2026, 02:47 PM
Security Audit — agent-trust-hub — heroui-native