@vibeapi/cli@0.0.139
CLI tool for managing AI coding assistants (Claude Code, Codex, OpenClaw, Hermes, AtomCode, Grok Build) with one unified config
AI Security Review
scanned 8d ago · by lpm-firewall-aiThe first-party plugin setup creates residual agent-extension risk because applying Claude Code configuration can install plugins without a separate confirmation. Credential forwarding is conditional on the configured destination, and the default is package-aligned; no concrete malicious behavior was established.
Decision evidence
public snapshot- Applying Claude Code configuration calls the plugin installer when a Claude executable is found; the setup flow reaches this configuration application.
- The installer adds a marketplace and installs plugins that are not already installed or recorded, without a separate plugin-level confirmation.
- The request helper reads the saved API key and attaches it as a bearer token to a POST resolved against the active base URL; the key is saved by the authentication command.
- The base URL can be set with --base and accepts arbitrary HTTP or HTTPS URLs; the packaged default is VibeAPI’s gateway.
- Telemetry is enabled by default unless its setting disables it. CLI commands record telemetry, and a separate doctor --report action can upload a diagnostic report.
- The configured default gateway belongs to the package’s VibeAPI brand, and telemetry has a disable setting.
- The inspected sources show official Claude plugin installation and conditional credential forwarding, but no malicious plugin payload or confirmed credential theft.
Source & flagged code
6 flaggedPackage source references child process execution.
dist/lib/tools/desktop-process.jsView on unpkg · L1Findings
Affected versions and remediation
This report applies to @vibeapi/cli@0.0.139.
See all versions of @vibeapi/cli for other recorded verdicts.
- Review the evidence and your use of @vibeapi/cli@0.0.139 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.