Warn by default; block when configured. Changes the user's broader AI-agent configuration without an explicit request, allowing Claude Code to launch the package's MCP server.
AI security review
scanned · by lpm-firewall-ai
The install hook automatically registers a bundled MCP by writing to the user's Claude Code configuration outside the package directory. This is install-time AI-agent control hijacking under the stated review rules.
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
npm install or upgrade, through the package's postinstall hook.
Impact
Changes the user's broader AI-agent configuration without an explicit request, allowing Claude Code to launch the package's MCP server.
Mechanism
The hook invokes the CLI to register a bundled MCP server, which writes its launch entry to ~/.claude.json.
Policy narrative
During installation, the postinstall hook registers the bundled dr-zendesk-kb MCP with Claude Code. The CLI writes a launch entry to the user's ~/.claude.json, making this an automatic install-time change to broader AI-agent settings.
Avoid installing @datarails-ai/dr-cli@1.6.190. Remove it from direct dependencies and check your lockfile for transitive copies.
Choose an independently verified alternative or release. This report does not establish that other versions are safe.
If this version ran, check your AI agents' configuration and instruction files (the files this report lists) for changes you did not make, and remove them. This finding does not by itself establish malware intent.