@whalent/agent-core@0.3.497
Core runtime for Whalent Agent
AI Security Review
scanned 14d ago · by lpm-firewall-aiThe published entry point is heavily obfuscated, and static inspection did not resolve its full behavior. The inspected network code shows DNS resolver use, but no concrete attack chain; warn rather than block.
Decision evidence
public snapshot- The manifest exposes a bundled executable as both the package entry point and CLI.
- The executable bundle begins with a large obfuscated string-table rotation loop, which leaves its behavior difficult to verify.
- The MCP bundle includes DNS-over-HTTPS endpoint selection and environment-based configuration; the inspected excerpt does not show credential collection or an exfiltration sink.
- The manifest declares no install lifecycle scripts.
- The inspected network code uses public DNS-over-HTTPS resolvers and does not establish an attack.
Source & flagged code
7 flagged · loading sourceSource contains an obfuscated payload loader that reconstructs and executes hidden code.
dist/mcp.cjsView on unpkg · L1A bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.
dist/mcp.cjsView on unpkgFindings
Affected versions and remediation
This report applies to @whalent/agent-core@0.3.497.
See all versions of @whalent/agent-core for other recorded verdicts.
- Review the evidence and your use of @whalent/agent-core@0.3.497 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.