@viraatdas/rudder@3.1.8
⚠ Under reviewA 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 install hook appears limited to installing the declared Jujutsu prerequisite, and the inspected usage request sends a saved Codex token to chatgpt.com. Cloud snapshot code stages agent credentials and configuration, but the inspected source does not resolve its full destination and consent path; further review is needed before deciding whether that behavior is benign.
Decision evidence
public snapshot- package.json declares a postinstall hook; it checks for Jujutsu and, when absent, attempts installation through Homebrew or Cargo.
- Cloud functionality reads local agent credential files and stages Claude credentials and agent configuration into a home snapshot; the inspected excerpts do not establish the full upload destination or consent path.
- The usage code sends a saved Codex access token to the ChatGPT usage endpoint, and the board exposes quota data through a local loopback server.
- The install hook contains no inspected project mutation or unrelated script-approval changes.
- The inspected credential-quota request targets chatgpt.com, the service associated with the saved Codex token.
Source & flagged code
18 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@3.1.8.
See all versions of @viraatdas/rudder for other recorded verdicts.
- Review the evidence and your use of @viraatdas/rudder@3.1.8 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.