registry / @whalent/agent-core / 0.3.501
@whalent/agent-core@0.3.501
Core runtime for Whalent Agent
AI Security Review
scanned 12d ago · by lpm-firewall-aiThe source is heavily obfuscated, leaving meaningful behavior difficult to verify, so a warning is appropriate. The inspected excerpts show configurable, caller-triggered platform and skill operations, but do not establish concrete malicious behavior.
Static reason
One or more suspicious static signals were detected.
Trigger
Running the CLI directly and invoking its MCP functionality.
Impact
Could make authenticated platform requests and alter skill directories through its MCP functionality; malicious use was not established.
Mechanism
It uses a configured token for platform API requests and has code that writes skill files.
AI rationale
The source is heavily obfuscated, leaving meaningful behavior difficult to verify, so a warning is appropriate. The inspected excerpts show configurable, caller-triggered platform and skill operations, but do not establish concrete malicious behavior.
Evidence
dist/mcp.cjs
Network endpoints1
memory.whalent.com
Decision evidence
public snapshotAI called this Suspicious at 77.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- The MCP entry is heavily obfuscated, limiting reliable static verification of its behavior.
- The CLI accepts a configurable platform URL and token, then uses the token in platform API requests.
- The MCP source can create directories and write skill files.
Evidence against
- package.json declares no install lifecycle scripts.
- The CLI configuration is assembled when the executable is run directly.
- The inspected source does not establish that the configurable endpoint is package-controlled or that credentials are sent without caller configuration.
Behavioral surface
CryptoDynamicRequireEnvironmentVarsEvalFilesystemNetwork
HighEntropyStringsMinifiedObfuscatedUrlStrings
NoLicense
Source & flagged code
7 flagged · loading sourcedist/mcp.cjsView file
1#!/usr/bin/env node
L2: 'use strict';var a0_0xeb5671=a0_0xf006;(function(_0x55bcfb,_0x323dcf){var _0x168c1e=a0_0xf006,_0x1d4775=_0x55bcfb();while(!![]){try{var _0x14ff66=parseInt(_0x168c1e(0x2e3))/0x1+par...
L3: || (${_0x4b8e48} == "string" && ${_0x1ea01c} && ${_0x1ea01c} == +${_0x1ea01c})`)[_0x26e8ee(0x1e6f)](_0x4bf79d,(0x0,_0x3dcaa5['_'])`+${_0x1ea01c}`);return;case _0x26e8ee(0x23cb):_0x...
High
Obfuscated Payload Loader
Source contains an obfuscated payload loader that reconstructs and executes hidden code.
dist/mcp.cjsView on unpkg · L1•stage = ast_semantic_analysis; reason = ast_path_work_budget_exceeded; limitedFiles = 1
High
Semantic Analysis Limited
A bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.
dist/mcp.cjsView on unpkg4|| (${_0x4b8e48} === "string" && ${_0x1ea01c} && ${_0x1ea01c} == +${_0x1ea01c} && !(${_0x1ea01c} % 1))`)[_0x26e8ee(0x1e6f)](_0x4bf79d,(0x0,_0x3dcaa5['_'])`+${_0x1ea01c}`);return;ca...
L5: || ${_0x4b8e48} === "boolean" || ${_0x1ea01c} === null`)[_0x26e8ee(0x1e6f)](_0x4bf79d,(0x0,_0x3dcaa5['_'])`[${_0x1ea01c}]`);}}}function _0x2bfc3c({gen:_0x27845a,parentData:_0x3154a...
L6: missingProperty: ${_0x11aabc},
Medium
Findings
3 High7 Medium5 Low
HighObfuscated Payload Loaderdist/mcp.cjs
HighObfuscated
HighSemantic Analysis Limiteddist/mcp.cjs
MediumDynamic Requiredist/mcp.cjs
MediumNetwork
MediumEnvironment Vars
MediumShips Build Helperdist/jupyter_sidecar.py
MediumOversized Source Filedist/index.cjs
MediumOversized Cli Entrypointdist/index.cjs
MediumStructural Risk Force Deep Review
LowEvaldist/mcp.cjs
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
LowNo License
Affected versions and remediation
This report applies to @whalent/agent-core@0.3.501.
See all versions of @whalent/agent-core for other recorded verdicts.
- Review the evidence and your use of @whalent/agent-core@0.3.501 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.