devez-vibe@1.9.30
Stable terminal UI for Codex and Claude Agent SDK
AI Security Review
scanned 25d ago · by lpm-firewall-aiAutomatic installation into global AI-agent skill roots is a concrete extension lifecycle risk. The inspected source does not establish the additional attack behavior required for blocking.
Decision evidence
public snapshot- The package declares a post-install hook that runs the skill installer and Python dependency setup.
- The skill installer targets the global Codex and Claude skill directories under the user configuration locations.
- That install-time script copies bundled skill trees into those agent directories and removes entries absent from the bundled source.
- The post-install setup invokes user-scoped pip installation.
- The bridge can write Claude permission settings when its permission-update operation is invoked.
- The inspected manifest has no self-dependency.
- No credential collection or exfiltration sink was identified in the inspected locations.
Source & flagged code
9 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 creates an unconsented AI-agent control surface through install-time mutation or a default unauthenticated remote skill channel.
Findings
Affected versions and remediation
This report applies to devez-vibe@1.9.30.
See all versions of devez-vibe for other recorded verdicts.
- Review the evidence and your use of devez-vibe@1.9.30 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.