@cloop/runner@0.3.38
Make your computer and your agents Cloop machines: device login, the machine daemon, the Cloop MCP server, and Claude Code hooks.
AI Security Review
scanned 3d ago · by lpm-firewall-aiThe reviewed process execution and network access support the package's stated coding-agent connection workflow. The install is triggered by `cloop connect`, announced to the user, and uses Anthropic's official installer host.
Decision evidence
public snapshot- The package describes its purpose as connecting computers and coding agents to Cloop.
- The CLI runs the connect flow only when the user invokes `cloop connect`.
- When that flow finds no suitable coding agent, it announces that it will install Claude Code using Anthropic's installer.
- The installer command uses Anthropic's official `claude.ai` install script.
- The install command downloads and runs a shell script, but it is part of the user-invoked connect flow and is announced before running.
- The scanner's child-process and network findings are consistent with installing and managing coding agents; no unrelated data collection or attack behavior was identified.
Source & flagged code
5 flaggedPackage source references child process execution.
dist/chunk-SBYSVZ7J.jsView on unpkg · L84Source gates dangerous network, credential, or execution behavior behind CI, host, platform, time, or geo fingerprint checks.
dist/chunk-3WRLILTX.jsView on unpkg · L48Findings
Affected versions and remediation
This report applies to @cloop/runner@0.3.38.
See all versions of @cloop/runner for other recorded verdicts.
- Review the evidence and your use of @cloop/runner@0.3.38 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.