@whalent/agent@0.3.410
Stable wrapper for Whalent Agent core runtime
AI Security Review
scanned 2mo ago · by lpm-firewall-aiThis release automatically updates a global executable at runtime from opaque code, which is unsafe and not adequately auditable. There is no install lifecycle hook or confirmed direct malicious action in the inspected package.
Decision evidence
public snapshot- The CLI entrypoint is heavily obfuscated, including a runtime string decoder, which prevents transparent review of its behavior.
- Every normal CLI invocation calls an automatic wrapper-update routine before starting the supervised core.
- The update code uses npm global installation for the package name and selected version, allowing runtime replacement of the globally installed tool.
- package.json has no preinstall, install, or postinstall hook; prepublishOnly is not run for normal consumers.
- No concrete credential theft, data exfiltration, destructive command, or foreign AI-agent configuration write was confirmed by source inspection.
- The bundled Python sidecar is a Jupyter kernel protocol helper rather than an install-time launcher.
Source & flagged code
5 flagged · loading sourceSource contains an obfuscated payload loader that reconstructs and executes hidden code.
dist/index.cjsView on unpkg · L1Package source references dynamic require/import behavior.
dist/index.cjsView on unpkg · L1Findings
Affected versions and remediation
This report applies to @whalent/agent@0.3.410.
See all versions of @whalent/agent for other recorded verdicts.
- Review the evidence and your use of @whalent/agent@0.3.410 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.