flavor-code@1.4.1
<p align="center"><a href="./README.md">English</a> | <b><a href="./README.zh-CN.md">简体中文</a></b></p>
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThe package has broad but package-aligned interactive coding capabilities and no registry install lifecycle hook. The only confirmed reporting path is opt-in and defaults to localhost, so the scanner's claimed credential redirection and automatic exfiltration are not supported by inspected source.
Decision evidence
public snapshot- The desktop runtime includes an opt-in incident reporter that can send failed tool input and workspace metadata to a configured webhook.
- The application exposes user-invoked terminal capabilities, so command execution is a real runtime capability.
- The command helper runs a specified executable with supplied arguments, timeout, working-directory, and environment options.
- The CLI configures Node crash diagnostics to write reports and heap snapshots next to the working directory.
- package.json has no preinstall, install, or postinstall hook.
- No fixed unofficial credential gateway or command-output exfiltration endpoint was found in the inspected high-risk files.
- The reporter is disabled unless explicitly enabled; its default endpoint is localhost.
- The command and network features are part of an interactive coding application and include permission controls.
Source & flagged code
11 flagged · loading sourcePackage source references child process execution.
dist/chunk-ZJCAQXA7.jsView on unpkg · L4Findings
Affected versions and remediation
This report applies to flavor-code@1.4.1.
See all versions of flavor-code for other recorded verdicts.
- Review the evidence and your use of flavor-code@1.4.1 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.