Static source confirms automatic first-party agent skill and MCP setup during npm installation, a guarded agent extension lifecycle risk. The inspected credential flow is user-invoked and constrained to HTTPS Datarails hosts, so the scanner’s credential-exfiltration hint is not supported by these source paths.
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 npm package runs its postinstall hook.
Impact
AI assistant sessions may load package-provided skills and launch the registered MCP server without a separate setup confirmation.
Mechanism
The hook installs bundled skills and registers dr-zendesk-kb with Claude Code; Gemini skill configuration is also updated when its directory exists.
Evidence
package.jsonscripts/postinstall.mjsdist/cli.js~/.claude/skills/~/.claude.json~/.gemini/skills/~/.gemini/GEMINI.md