create-framework-component

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute pnpm generate within the packages/docs directory to rebuild documentation after component scaffolding. This is a standard project-specific development command.
  • [INDIRECT_PROMPT_INJECTION]: The skill identifies a surface for indirect prompt injection by processing user-provided component names and types to generate files and directory structures.
  • Ingestion points: User input for kebab-case names, PascalCase namespaces, and component types provided via interactive prompts (SKILL.md).
  • Boundary markers: The skill requires a manual confirmation step where the planned file tree and API structure must be reviewed and approved by the user before files are created.
  • Capability inventory: The skill performs file system writes and local shell command execution via pnpm (SKILL.md).
  • Sanitization: No specific delimiters or validation rules for user input are mentioned; however, the structured confirmation flow mitigates the risk of unintended behavior.
  • [EXTERNAL_DOWNLOADS]: References external documentation from zagjs.com, which is a well-known service for UI component state machines. This is documented as a legitimate implementation reference.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 07:18 PM