pennyrouter@0.3.66
Install and manage PennyRouter local coding-agent integrations.
AI Security Review
scanned 6d ago · by lpm-firewall-aiSource inspection found no install lifecycle hook or concrete attack behavior. Credential forwarding is part of gateway configuration, and the flagged agent configuration behavior is reached through the CLI integration flow.
Decision evidence
public snapshot- The manifest exposes a CLI and has no install lifecycle hook; it also ships a standalone web demo.
- Claude credential reads occur in an integration flow gated on subscription, gateway, and API key; credentials are stored for the selected gateway.
- The CLI passes upstream credentials to that gateway, and its defaults and configuration support PennyRouter gateway setup.
- Agent configuration changes are runtime integration behavior; the flagged test paths contain test setup or local path helpers.
- The entrypoint delegates to the CLI, while CodeGPT prints manual setup instructions instead of writing IDE settings.
- Inspection found no confirmed attack surface: no unconsented install-time action, unrelated credential recipient, or malicious agent configuration change.
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@0.3.66.
See all versions of pennyrouter for other recorded verdicts.
- Review the evidence and your use of pennyrouter@0.3.66 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.