pennyrouter@0.3.43
Install and manage PennyRouter local coding-agent integrations.
AI Security Review
scanned 11d ago · by lpm-firewall-aiThe package provides user-invoked setup that writes agent configuration and registers an MCP server, so the capability merits a warning. Inspected source does not establish malicious behavior or automatic npm-install execution.
Decision evidence
public snapshot- package.json exposes a command-line entry point and defines only start and check scripts, with no npm install lifecycle hook.
- The explicit install command configures selected harnesses; Claude Code setup writes a user skill and launch settings.
- The installer registers the Threads MCP server unless the user passes --no-mcp; registration updates supported harness configuration.
- The remaining flagged locations inspected concern test fixtures, local session or transcript reads, and the standalone demo; they do not establish an attack.
- The CodeGPT integration shows manual setup instructions and avoids modifying IDE settings.
- No source inspected establishes automatic execution during npm installation, credential theft, or an unrelated remote payload.
Source & flagged code
10 flagged · loading sourcePackage source references child process execution.
standalone/web-demo/server.mjsView on unpkg · L3A single source file combines environment access, network access, and code or shell execution; review context before blocking.
standalone/web-demo/server.mjsView on unpkg · L304Findings
Affected versions and remediation
This report applies to pennyrouter@0.3.43.
See all versions of pennyrouter for other recorded verdicts.
- Review the evidence and your use of pennyrouter@0.3.43 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.