@tiens.nguyen/gu-cli@1.0.837
Gu CLI — the gu terminal plus the local worker that runs agent / OCR / PDF / embedding jobs on your Mac (Ollama / MLX / OpenAI-compatible).
AI Security Review
scanned 7d ago · by lpm-firewall-aiThe scanner’s credential-redirection claim is not supported by the inspected source: login is explicit and pairs with the package’s worker API, while worker keys authenticate requests to the configured API. The postinstall is limited to PATH setup, and the inspected agent configuration paths include approval gates.
Decision evidence
public snapshot- The postinstall runs install-path.mjs to help users find the CLI by adding its global binary directory to shell PATH.
- Login is an explicit CLI action; device pairing uses the package’s worker API, and its flow asks the user to approve the pairing in a browser.
- Worker keys are sent to the configured worker API as authentication for worker routes; the inspected source does not show transfer to an unrelated recipient.
- Project hooks are run only when approved, and the agent hook settings location is a path helper rather than an automatic install-time mutation.
- Project MCP server approval is an explicit configuration operation; the inspected MCP location is a path helper.
- The inspected MongoDB, provider URL, proxy, and agent-stop code implements local service probing, provider URL vetting, per-request proxying, and stopping ended jobs.
Source & flagged code
14 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgPackage source references child process execution.
Findings
Affected versions and remediation
This report applies to @tiens.nguyen/gu-cli@1.0.837.
See all versions of @tiens.nguyen/gu-cli for other recorded verdicts.
- Review the evidence and your use of @tiens.nguyen/gu-cli@1.0.837 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.