@metaphi-ai/hum@0.2.14
Hum: a self-improving coding agent for your terminal.
AI Security Review
scanned 14d ago · by lpm-firewall-aiThe flagged lifecycle and credential access are present, but inspected source ties them to bootstrapping and running the declared terminal coding agent. No concrete malicious behavior was established.
Decision evidence
public snapshot- The manifest identifies Hum as a terminal coding agent and declares its launcher and postinstall hook.
- Postinstall runs the core installer only when the matching version is not already installed.
- The installer bootstraps its Python runtime through uv, consistent with the declared agent package.
- The CLI uses the Hum service endpoint by default and reads credentials from its own Hum directory; the inspected code showed no browser credential collection or unrelated recipient.
- The postinstall hook can execute a downloaded uv installation script through a shell or PowerShell.
- The CLI reads a Hum access token from ~/.hum/credentials.json.
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@0.2.14.
See all versions of @metaphi-ai/hum for other recorded verdicts.
- Review the evidence and your use of @metaphi-ai/hum@0.2.14 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.