devez-vibe@1.7.36
Stable terminal UI for Codex and Claude Agent SDK
AI Security Review
scanned 2mo ago · by lpm-firewall-aiThe package has a concrete automatic AI-agent extension installation surface, so it should be warned. The source limits skill use to explicit invocation and contains no confirmed exfiltration or remote payload behavior.
Decision evidence
public snapshot- The automatic postinstall hook runs an installer.
- The installer copies and prunes a package-supplied skill in both Codex and Claude user skill directories.
- The installed skill can direct a host to launch Codex CLI subprocesses when explicitly invoked.
- The skill declares that it is used only when the user explicitly invokes it.
- No source evidence shows network exfiltration, credential harvesting, or remote payload download.
- The Claude bridge removes Anthropic credential variables before launching its SDK environment.
Source & flagged code
7 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgThe automatic postinstall hook runs an installer.
package.jsonView on unpkg · L40Findings
Affected versions and remediation
This report applies to devez-vibe@1.7.36.
See all versions of devez-vibe for other recorded verdicts.
- Review the evidence and your use of devez-vibe@1.7.36 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.