@exla/rudder@3.9.2
⚠ 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 1d ago · by lpm-firewall-aiThe inspected sources show credential use directed to the corresponding account services and a setup-oriented install hook. The improvement collection flow remains materially unresolved, so this review cannot determine whether it exposes project session data.
Decision evidence
public snapshot- The manifest runs a postinstall setup script.
- The improvement feature reads registered projects and builds session records, and it includes a scheduled-job component.
- The cloud code reads a Codex identity claim, while the usage code reads a saved access token for a ChatGPT usage request.
- The improvement collection and processing flow remains unresolved in the inspected source.
- The postinstall script skips when its environment flag is set and returns when run as root.
- The inspected Codex usage request targets ChatGPT’s usage endpoint.
Source & flagged code
24 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgInstall-time source downloads a native archive from a fixed external host without transport verification, extracts it, and installs an executable payload.
package.jsonView on unpkg · L1Findings
Affected versions and remediation
This report applies to @exla/rudder@3.9.2.
See all versions of @exla/rudder for other recorded verdicts.
- Review the evidence and your use of @exla/rudder@3.9.2 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.