@metaphi-ai/hum@0.1.10
Hum (हम): a self-improving coding agent for your terminal.
AI Security Review
scanned 14d ago · by lpm-firewall-aiThe source confirms an automatic postinstall that may execute remotely fetched installer code and install the agent engine. Since this behavior is package-aligned and no concrete malicious action was found, classify it as a warning for conditional install-time code execution rather than malware.
Decision evidence
public snapshot- The manifest runs scripts/postinstall.js automatically during installation.
- The postinstall hook downloads and runs Astral's installer through a shell when uv is missing.
- The hook installs the matching hum-cli engine into the user's Hum home directory.
- The launcher imports its bundled CLI, which starts the engine as the terminal agent.
- The hook's engine installation matches the package's terminal coding-agent purpose.
- The inspected source showed no browser credential collection or unrelated data exfiltration.
- The bundle's network and environment matches were not evidence of credential theft; its active engine communication uses localhost.
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@0.1.10.
See all versions of @metaphi-ai/hum for other recorded verdicts.
- Review the evidence and your use of @metaphi-ai/hum@0.1.10 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.