The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from multiple sources which could be used to influence agent behavior or project files.\n
Ingestion points: User instructions, Figma design nodes, source copy, and external component metadata from the PrimeUI registry.\n
Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands in the ingested data.\n
Capability inventory: The agent is authorized to write to the local file system, execute shell commands (npx, build scripts), and interact with external APIs via MCP tools.\n
Sanitization: The skill implements a validation step for component properties using component_props_validate before materializing components into the project.\n- [EXTERNAL_DOWNLOADS]: The skill relies on external CLI tools and MCP servers fetched dynamically from the npm registry.\n
Evidence: Instructions to run npx @primeuicom/cli setup and npx @primeuicom/mcp@latest --health verify the presence and execution of external packages.\n- [COMMAND_EXECUTION]: The skill facilitates the execution of shell commands for project initialization, health checks, and verification.\n
Evidence: The skill explicitly mentions running npx commands and standard project scripts such as typecheck, build, and dev to verify local edits. The setup command also includes an --ai-preset <agent> flag which interpolates agent-supplied values into the command line.

primeui-page-builder