vibe-coding-mcp-setup
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references several official and well-known MCP server packages from the
@modelcontextprotocoland@playwrightorganizations on NPM. These are established sources for MCP infrastructure. - Evidence includes:
npx -y @modelcontextprotocol/server-postgres,npx -y @modelcontextprotocol/server-github,npx -y @playwright/mcp, etc. - [COMMAND_EXECUTION]: The skill provides configuration blocks that use
npxto execute these MCP servers. This is standard behavior for configuring the Model Context Protocol in AI agent environments. - [SAFE]: The skill explicitly emphasizes security best practices throughout its instructions.
- It advises users to use read-only database users and avoid production credentials.
- It includes a dedicated Security Checklist reminding users to use environment variables for secrets, scope filesystem access to project directories, and use fine-grained GitHub tokens.
- It instructs the agent to help users get credentials securely rather than requesting them directly.
Audit Metadata