The unprompted postinstall changes user-level assistant configuration and registers an MCP server, leaving a concrete agent integration risk. Its inspected scope is package-owned help-center setup, so the evidence supports a warning rather than a malicious verdict.
Static reason
High-risk behavior combination matched malicious policy.; source matched previously finalized malicious package; routed for review; source fingerprint signature matched known malicious package; routed for review
Trigger
Installing the package through npm runs its postinstall hook.
Impact
The assistant may load the installed skills and expose the registered help-center tools in later sessions.
Mechanism
The hook invokes the CLI to copy embedded skills and register dr-zendesk-kb in the user's Claude configuration.
Evidence
package.jsonscripts/postinstall.mjsdist/cli.jsdist/zendeskKb-JPSHVOPH.js~/.claude.json~/.claude/skills/~/.gemini/skills/~/.gemini/GEMINI.md