@su-record/vibe@4.2.21
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 4d ago · by lpm-firewall-aiThe global postinstall automatically configures this package’s Claude or Codex integration, creating residual lifecycle risk for agent setup. The inspected behavior is package-owned integration setup, with no confirmed broader attack behavior; downgrade to a warning.
Decision evidence
public snapshot- package.json defines a postinstall hook; scripts/postinstall.js runs setup for global npm installs unless VIBE_SKIP_SETUP is set.
- The setup routine processes detected client homes, registers the package’s Claude or Codex plugin, and falls back to writing home-level integration files.
- The package builds agent hooks for session, stop, and tool-use events. The inspected plugin installer assembles the local plugin tree and the Claude registration code checks installed plugin metadata.
- The cited plugin and registration locations were inspected as required; they sit within the first-party local plugin setup flow.
- The inspected install path is limited to setting up this package’s agent integrations; no unrelated credential transfer or remote payload execution was identified there.
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@4.2.21.
See all versions of @su-record/vibe for other recorded verdicts.
- Review the evidence and your use of @su-record/vibe@4.2.21 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.