The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill provides an installation command that downloads a script from the vendor's official domain (heroui.com) and executes it via bash. This is a standard installation pattern for this vendor.
[EXTERNAL_DOWNLOADS]: Utility scripts in the scripts/ directory fetch component documentation, source code, and styling information from vendor-managed endpoints including mcp-api.heroui.com, heroui.com, and official GitHub repositories.
[COMMAND_EXECUTION]: Instructs the agent to execute local JavaScript files using Node.js to manage documentation and component retrieval tasks.
[DATA_EXFILTRATION]: No instances of unauthorized data access or exfiltration were detected. All network operations are directed towards well-known vendor infrastructure and public repositories.
[PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it ingests data from remote sources.
Ingestion points: Documentation and source code are fetched from heroui.com and GitHub in scripts/get_component_docs.mjs, scripts/get_docs.mjs, and scripts/get_source.mjs.
Boundary markers: Absent; fetched content is displayed directly to the agent without delimiters.
Capability inventory: The skill provides commands for local script execution and installation tasks.
Sanitization: No evidence of content sanitization or validation before the fetched data is processed by the agent.

heroui-react