@metaphi-ai/hum@0.1.24
Hum (हम): a self-improving coding agent for your terminal.
AI Security Review
scanned 14d ago · by lpm-firewall-aiSource inspection shows a postinstall hook that provisions the package's matching engine and a CLI that invokes it on user commands. The reviewed evidence does not substantiate the scanner's browser-credential-theft claim or another malicious attack.
Decision evidence
public snapshot- The manifest wires `postinstall` to a package-owned installer and exposes the Hum CLI.
- The installer checks for the matching engine version, then installs `uv` and the matching `hum-cli` Python environment when needed.
- The bundle defaults its update distribution host to `storage.googleapis.com/hum-dist`.
- The bundle runs the engine when a user invokes one of its engine commands.
- Inspected installer behavior is consistent with setting up the package's own engine under the Hum home directory.
- No browser credential collection, secret forwarding, or unrelated exfiltration behavior was established in the inspected source.
Source & flagged code
10 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@0.1.24.
See all versions of @metaphi-ai/hum for other recorded verdicts.
- Review the evidence and your use of @metaphi-ai/hum@0.1.24 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.