datahub-mfe-create-app
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill performs shell commands including directory creation,
npm install, andnpm startto initialize the project and launch the development server. - [INDIRECT_PROMPT_INJECTION]: The skill incorporates user-provided application names and descriptions into the generated project files.
- Ingestion points: User input collected via the
AskQuestiontool forapp_nameandapp_descinSKILL.md. - Boundary markers: Absent; user strings are directly substituted into template placeholders.
- Capability inventory: The agent creates directories/files and executes
npmcommands (install,start) inSKILL.md. - Sanitization: No explicit sanitization or validation of the user-provided strings is performed before they are written to
package.json,App.tsx, andindex.html.
Audit Metadata