project-setup-info-local

Pass

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [Command Execution]: The skill initiates project setup using npx to execute scaffolding packages such as create-next-app and create-vite. This is a standard method for bootstrapping modern web and extension projects.
  • [External Data Ingestion]: Documentation is retrieved from external official sources to guide MCP server setup. This represents a potential surface for indirect prompt injection. Ingestion points: fetch_webpage tool retrieves content from modelcontextprotocol.io. Boundary markers: None explicitly defined for the external text to distinguish it from system instructions. Capability inventory: File system access for creating mcp.json and updating .github files, and VS Code command execution. Sanitization: Not explicitly performed on the fetched text.
  • [VS Code Integration]: The skill uses official VS Code commands to initialize Python environments and install relevant extensions, which represents normal integration with the development environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 27, 2026, 07:43 PM