pennyrouter@0.3.50
Install and manage PennyRouter local coding-agent integrations.
AI Security Review
scanned 11d ago · by lpm-firewall-aiSource inspection found a real agent configuration capability: install and update commands write local integration settings, with update registering the package-owned MCP server unless skipped. This supports a warning for user-invoked agent capability, not a malicious or install-hook verdict.
Decision evidence
public snapshot- package.json defines start and check scripts but no install lifecycle hook.
- The package entrypoint invokes its CLI; installation proceeds through the explicit install command.
- The update path registers the package’s MCP server for installed harnesses unless the user supplies --no-mcp; this mutates agent configuration on an explicit update.
- The MCP server entry launches PennyRouter itself, and the registration writer skips absent Claude configuration rather than creating it.
- The installer writes a Claude Code skill and launch settings as part of harness setup.
- The inspected source does not establish credential theft, an unrelated network recipient, destructive behavior, or a hidden install-time trigger.
- The statusline integration preserves an existing user command for restoration, and the package exposes an uninstall path.
Source & flagged code
7 flagged · loading sourceSource writes installer persistence such as shell profile or service configuration.
src/launch.test.jsView on unpkg · L1Findings
Affected versions and remediation
This report applies to pennyrouter@0.3.50.
See all versions of pennyrouter for other recorded verdicts.
- Review the evidence and your use of pennyrouter@0.3.50 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.