@viraatdas/rudder@2.14.24
⚠ Under reviewA Claude Code-style terminal app for running coding agents with worktree-isolated runs.
AI Security Review
scanned 15d ago · by lpm-firewall-aiInspected source shows cloud workspace credential staging and snapshot upload, plus a postinstall dependency setup, but does not establish a concrete malicious chain. The required clean citation coverage exceeds the eight-citation limit, so the review remains unresolved.
Decision evidence
public snapshot- The manifest defines a postinstall hook and a CLI entrypoint.
- The postinstall script checks for jj and may install it with brew or cargo.
- The cloud module reads Claude credentials from macOS Keychain and stages them in a workspace home.
- Cloud attach builds a migration plan and may upload a workspace snapshot.
- Handoff and model helpers read Claude conversation directories and a Codex model cache.
- The inspected source shows cloud credential staging and workspace upload within cloud workspace functionality; the inspected excerpts do not establish an unrelated recipient or automatic npm-triggered credential transfer.
- The inspected manifest and postinstall excerpt describe jj setup; the excerpt does not show broad project or agent-configuration mutation.
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@2.14.24.
See all versions of @viraatdas/rudder for other recorded verdicts.
- Review the evidence and your use of @viraatdas/rudder@2.14.24 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.