@metaphi-ai/hum@0.2.23
Hum: a self-improving coding agent for your terminal.
AI Security Review
scanned 14d ago · by lpm-firewall-aiThe postinstall and subprocess behavior serve the package’s own versioned core setup, and the CLI delegates work to that core when launched. Although the bootstrap executes an upstream installer, inspected source did not show a concrete attack chain or credential exfiltration.
Decision evidence
public snapshot- package.json links the CLI to a postinstall script and the `hum` launcher.
- dist/install.mjs uses the upstream Astral installer to bootstrap uv, then installs Hum’s Python core into a version-specific directory.
- dist/cli.mjs starts the core when a session is invoked; the inspected source did not establish credential harvesting or an unrelated exfiltration path.
- The postinstall runs a shell command that pipes the Astral-hosted uv installer into the shell; this is automatic bootstrap behavior but is consistent with setting up Hum’s core.
Source & flagged code
9 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgInstall-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkgFindings
Affected versions and remediation
This report applies to @metaphi-ai/hum@0.2.23.
See all versions of @metaphi-ai/hum for other recorded verdicts.
- Review the evidence and your use of @metaphi-ai/hum@0.2.23 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.