@mutmutco/claude-plugin@4.5.114
MMI workflow skills and org gates delivery.
AI Security Review
scanned 5d ago · by lpm-firewall-aiThe manifest has no lifecycle hook, and the flagged agent integration location is only part of a path classifier. However, Hub authentication forwards an existing GitHub token to an unrestricted MMI_HUB_URL override, leaving a concrete conditional credential exposure risk.
Decision evidence
public snapshot- package.json declares no lifecycle scripts.
- The CLI lets MMI_HUB_URL override its default Hub endpoint.
- GitHub credentials can come from environment variables or gh auth token.
- On a cache miss, Hub authentication posts the GitHub token as a bearer credential to the selected endpoint’s /auth/session route.
- Hub API callers invoke this authentication helper; the selected endpoint has no destination allowlist.
- The flagged agent path match is part of a path classifier, not an agent settings write.
- The default Hub endpoint is fixed in the CLI, and the credential exchange is used to obtain a Hub session.
- The manifest has no install hook, and the inspected code does not establish credential misuse by the default endpoint.
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/claude-plugin@4.5.114.
See all versions of @mutmutco/claude-plugin for other recorded verdicts.
- Review the evidence and your use of @mutmutco/claude-plugin@4.5.114 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.