@reason-machines/cli-win32-arm64@0.2.460
⚠ Under reviewReason CLI native binary for win32-arm64
AI Security Review
scanned 3d ago · by lpm-firewall-aiThe manifest has no lifecycle hook, and the agent-directory reference in the inspected runtime is used to construct local scan paths. However, bounded analysis of the CLI executable reveals credential-gathering code without resolving its reachability or destination, and the native addon is unsupported by the available analysis; this behavior remains unresolved.
Decision evidence
public snapshot- The manifest identifies a Windows ARM64 native CLI package and contains no lifecycle hook.
- The sandbox runtime lists Claude command and agent directories as paths used during local scanning.
- The bounded executable analysis contains a helper that gathers provider credentials, but does not establish whether it runs or where those values are sent.
- The executable analysis warns that its source windows cannot establish loader reachability or rule out other native behavior; the platform addon could not be inspected as source.
- The inspected sandbox code uses the agent-directory list to build local scan paths.
- The manifest contains package metadata and bundled files but no install-time script.
Source & flagged code
6 flagged · loading sourcePackage source references child process execution.
bin/runtime/srt/sandbox-runtime.mjsView on unpkg · L60A single source file combines environment access, network access, and code or shell execution; review context before blocking.
Findings
Affected versions and remediation
This report applies to @reason-machines/cli-win32-arm64@0.2.460.
See all versions of @reason-machines/cli-win32-arm64 for other recorded verdicts.
- Review the evidence and your use of @reason-machines/cli-win32-arm64@0.2.460 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.