The postinstall hook changes user-level AI assistant instructions and configuration without requiring a separate setup command. The changes appear limited to this package’s own bundled skills and MCP server, so this is an agent extension lifecycle risk rather than evidence of credential theft or a broader hijack.
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; previous stored version diff introduced dangerous source
Trigger
npm install or upgrade, through the postinstall hook.
Impact
AI assistant configuration and loaded instructions are changed without a separate setup command; the package’s MCP server is also made available to Claude Code.
Mechanism
The hook runs `dr skill install` for the home directory and registers `dr-zendesk-kb` with Claude Code.
Evidence
scripts/postinstall.mjsdist/cli.js~/.claude/skills/<skill>/~/.gemini/skills/<skill>/~/.gemini/GEMINI.md~/.claude.json