skills/sehoon787/my-claude/mcp-setup/Gen Agent Trust Hub

mcp-setup

Pass

Audited by Gen Agent Trust Hub on Mar 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill orchestrates the setup of external tools by executing system commands. It uses 'claude mcp add' for server registration, 'npx' for Node.js-based servers, and 'docker run' for the GitHub MCP implementation. It also allows for arbitrary user-defined commands through the 'Custom' server option.
  • [EXTERNAL_DOWNLOADS]: The skill retrieves software from well-known registries and official repositories. This includes downloading packages from the npm registry (such as '@upstash/context7-mcp', 'exa-mcp-server', and '@modelcontextprotocol/server-filesystem') and pulling a container image from the GitHub Container Registry ('ghcr.io/github/github-mcp-server').
  • [CREDENTIALS_UNSAFE]: The skill handles sensitive configuration data by prompting the user for an Exa API key and a GitHub Personal Access Token. These credentials are provided by the user and then used as environment variables within the setup commands to enable service access.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 24, 2026, 07:43 AM
Security Audit — agent-trust-hub — mcp-setup