@mutmutco/codex-plugin@4.5.174
MMI workflow skills and org gates delivery.
AI Security Review
scanned 6d ago · by lpm-firewall-aiSource inspection found no lifecycle hooks in package.json and no confirmed attack behavior. The credential transfer is part of a Hub session exchange, and the flagged agent-configuration path is used during plugin repair.
Decision evidence
public snapshot- package.json declares the package identity and ships the CLI; it defines no lifecycle hooks.
- The CLI’s default Hub endpoint is package-aligned with the MMI Hub service.
- The CLI sends a GitHub token to that Hub’s session endpoint as an authentication exchange.
- The agent-integration location identifies Claude’s marketplace settings file; the heal flow reads its pins, runs plugin repair steps, then restores those pins.
- The launcher requires the sibling CLI bundle only when invoked; its version-only path prints the package version.
Source & flagged code
7 flagged · loading sourceSource sends credentials or rich application records to a package-controlled external receiver enabled by default.
cli/dist/main.cjsView on unpkg · L39Source appears to send environment or credential material to an external endpoint.
cli/dist/main.cjsView on unpkg · L39Findings
Affected versions and remediation
This report applies to @mutmutco/codex-plugin@4.5.174.
See all versions of @mutmutco/codex-plugin for other recorded verdicts.
- Review the evidence and your use of @mutmutco/codex-plugin@4.5.174 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.