mcp-setup

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt repeatedly instructs the agent to prompt for API tokens/passwords and embed them verbatim into generated config files and commands (e.g., GITHUB_TOKEN in config.json, curl Authorization header, POSTGRES_CONNECTION_STRING), which requires the LLM to handle and output secret values directly.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill's configs and setup scripts invoke npx to fetch and run remote MCP server packages at runtime (e.g., "npx -y @modelcontextprotocol/server-github", "npx -y @modelcontextprotocol/server-postgres", "npx -y @modelcontextprotocol/server-filesystem", "npx -y @modelcontextprotocol/server-brave-search" and other "@modelcontextprotocol/server-*" packages), which downloads and executes remote code and are required by the skill to provide MCP servers.