@mutmutco/codex-plugin@4.5.149
MMI workflow skills and org gates delivery.
AI Security Review
scanned 5d ago · by lpm-firewall-aiThe source supports a conditional credential-exposure risk because an environment override can redirect the GitHub token exchange to an unrestricted origin. The default is the package Hub, the agent-config match is a path classifier, and package.json has no lifecycle scripts; the evidence does not support blocking the release.
Decision evidence
public snapshot- The manifest declares no lifecycle scripts, so the token exchange is not triggered by package installation.
- The CLI reads a GitHub token from environment variables or the GitHub CLI.
- Hub authentication sends that token as a bearer credential to the configured Hub session endpoint. MMI_HUB_URL can replace the package's default Hub URL, and the inspected exchange has no origin allowlist.
- Hub client wrappers pass the token into this exchange for Hub-backed commands; whoami also calls the session helper.
- The reported Codex path match is a repository path classifier used to identify uncommittable agent-config paths, not a configuration write.
- Separate CLI telemetry is triggered by a CLI command and sends command, version, CI status, and operating system metadata to the default Hub.
Source & flagged code
7 flagged · loading sourceSource appears to send environment or credential material to an external endpoint.
cli/dist/main.cjsView on unpkg · L39This package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
cli/dist/main.cjsView on unpkgFindings
Affected versions and remediation
This report applies to @mutmutco/codex-plugin@4.5.149.
See all versions of @mutmutco/codex-plugin for other recorded verdicts.
- Review the evidence and your use of @mutmutco/codex-plugin@4.5.149 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.