@su-record/vibe@4.2.6
Your personal FDE in Claude and Codex. Understand the work, use the right tools, deliver and retain useful context. One entry: vibe.
AI Security Review
scanned 11d ago · by lpm-firewall-aiThe postinstall behavior creates a guarded lifecycle risk because it sets up this package’s own agent surfaces in detected client homes. The inspected source does not establish unconsented modification of an unrelated control surface or another concrete attack, so this warrants a warning rather than a block.
Decision evidence
public snapshot- package.json defines a postinstall hook that runs scripts/postinstall.js.
- The hook runs setup only for global installs, unless VIBE_SKIP_SETUP is set.
- Global setup installs surfaces for each detected client home.
- Those surfaces include the package card, skill, and hook; available client CLIs may register the package’s own plugin.
- The plugin installer assembles the package tree and writes its marketplace entry.
- The required register.js location reads the installed Claude plugin version.
- The lifecycle setup is limited to global installs and has an explicit skip variable.
- The setup registers this package’s own plugin and installs its own card, skill, and hook; no unrelated agent extension or credential transfer was identified.
Source & flagged code
18 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 unpkgPackage source references child process execution.
mcpb/server/index.jsView on unpkg · L10Findings
Affected versions and remediation
This report applies to @su-record/vibe@4.2.6.
See all versions of @su-record/vibe for other recorded verdicts.
- Review the evidence and your use of @su-record/vibe@4.2.6 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.