@alumbwe/anvil@1.0.42
A Terminal User Interface (TUI) package built with OpenTUI and React.
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThe package has substantial agent command-execution, global-installation, and telemetry capability, but it lacks an install-time trigger or a concrete malicious chain. Its risk is real enough to warn rather than mark clean.
Decision evidence
public snapshot- The command broker launches an executable, arguments, working directory, and environment supplied by a request.
- Running the CLI automatically installs Bun globally when it is absent.
- Authenticated session calls send a bearer token to the Anvil service.
- Production logging ships buffered records to an API endpoint by default.
- Bundled configuration provides an AI-agent orchestration surface.
- The manifest has no npm install, preinstall, or postinstall hook; its only lifecycle hook is prepublishOnly.
- The global installation is triggered only after a user runs the anvil command.
- The inspected network calls are named Anvil product endpoints or telemetry paths, not an opaque payload host.
- No source inspected showed credential harvesting, hidden persistence, self-dependency, or install-time project mutation.
Source & flagged code
16 flagged · loading sourcePackage source references dynamic require/import behavior.
bin/anvil.jsView on unpkg · L99Findings
Affected versions and remediation
This report applies to @alumbwe/anvil@1.0.42.
See all versions of @alumbwe/anvil for other recorded verdicts.
- Review the evidence and your use of @alumbwe/anvil@1.0.42 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.