@tiens.nguyen/gu-cli@1.0.867
⚠ Under reviewGu 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 1d ago · by lpm-firewall-aiInspection did not substantiate the scanner's credential redirection claim in the reviewed login path, and the inspected agent configuration excerpts did not show automatic writes. The selected openai runtime dependency remained unresolved, so manual review is appropriate rather than certifying the package as clean.
Decision evidence
public snapshot- package.json declares a postinstall hook that runs install-path.mjs; the inspected installer describes a PATH setup task.
- The login command uses device-login.mjs, whose default service is chat-api-v2.gucli.com; the worker key is sent to that API for a read-only settings check.
- hooks-config.mjs and mcp-config.mjs use project and user configuration paths; inspected excerpts show path definitions and file reads.
- package.json declares openai as a runtime dependency, and gu-cli.mjs imports it; the prepared dependency source was unresolved, leaving part of runtime behavior unverified.
- The inspected login caller allows an explicit API base and defaults to the package's hosted API; no unrelated credential recipient was established in the inspected path.
- The inspected hook and MCP excerpts do not show automatic agent configuration writes.
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@1.0.867.
See all versions of @tiens.nguyen/gu-cli for other recorded verdicts.
- Review the evidence and your use of @tiens.nguyen/gu-cli@1.0.867 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.