@viraatdas/rudder@3.1.16
A TUI for managing coding agents: run Claude Code, Codex and opencode side by side, each in its own isolated jj workspace.
AI Security Review
scanned 6d ago · by lpm-firewall-aiThe postinstall behavior is limited to setting up the package’s stated jj dependency. The credential paths inspected use the Codex token with its usage service or compare local identities through the Rudder cloud API; no concrete attack was established.
Decision evidence
public snapshot- The postinstall hook checks for jj and, if needed, installs the Jujutsu tool through Homebrew or Cargo.
- Codex quota code reads the saved access token and sends it to chatgpt.com for the Codex usage endpoint; the caller requests quota data.
- Cloud credential parity reads the Codex identity email and compares local identities with the Rudder cloud account through its API.
- The inspected agent-related paths read local Claude and Codex data; task summaries redact secret-like tokens, and the board listens on loopback.
- The remaining cited runtime excerpts cover scheduling, collection, backend handling, update checks, and feedback behavior; no concrete attack was established.
- The scanner similarity labels and parser limits do not establish malicious behavior in the inspected source.
Source & flagged code
21 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgInstall-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkgSource file is highly similar to a previously finalized malicious package; route for source-aware review.
dist/util.jsView on unpkgFindings
Affected versions and remediation
This report applies to @viraatdas/rudder@3.1.16.
See all versions of @viraatdas/rudder for other recorded verdicts.
- Review the evidence and your use of @viraatdas/rudder@3.1.16 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.