The Agent Skills Directory

[SAFE]: The skill follows secure practices by explicitly instructing the agent to verify the existence of the mcp-moz package via npm view before attempting any usage, noting that npx can execute remote code.
[SAFE]: Data access is scoped to specific configuration and data directories (e.g., ~/.claude-marketing/ and ${CLAUDE_PLUGIN_DATA}), following the principle of least privilege for file system operations.
[SAFE]: The skill clearly distinguishes between its capabilities and external data sources (GSC, Rank-trackers), preventing the agent from making false claims about data it cannot verify, such as AI Overview citation exports from GSC.

rank-monitor