The postinstall hook automatically installs the package's own skills and registers its MCP server with Claude Code, an agent-extension lifecycle risk. The reviewed install path does not establish credential theft or unrelated network activity, so this warrants a warning rather than a publish block.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
npm install, through the postinstall lifecycle hook.
Impact
The package changes which skills and MCP server are available to assistant sessions without a separate user setup command.
Mechanism
The hook invokes `dr skill install` and `dr mcp register dr-zendesk-kb`; these write bundled skill files and a Claude MCP configuration entry.
Evidence
package.jsonscripts/postinstall.mjsdist/cli.js~/.claude/skills/<name>/~/.gemini/skills/<name>/~/.gemini/GEMINI.md~/.claude.json