registry / @whalent/agent / 0.3.459
@whalent/agent@0.3.459
Stable wrapper for Whalent Agent core runtime
AI Security Review
scanned 1mo ago · by lpm-firewall-aiInstall does not execute package code, and the runtime host and sibling dependency match the stated Whalent agent CLI. Obfuscated process and Claude config helpers are part of that user-started agent, not an install hook or a hidden exfiltration chain.
Static reason
One or more suspicious static signals were detected.
Trigger
A person runs the whalent command after install. npm install does not run a lifecycle hook.
Impact
No unconsented install mutation, credential theft, or remote payload execution was established from the inspected source.
Mechanism
The command starts the bundled agent runtime, which talks to the Whalent Memory host and can manage configured local workers.
AI rationale
Install does not execute package code, and the runtime host and sibling dependency match the stated Whalent agent CLI. Obfuscated process and Claude config helpers are part of that user-started agent, not an install hook or a hidden exfiltration chain.
Decision evidence
public snapshotAI called this Clean at 74.0% confidence as Benign with medium false-positive risk.
Evidence for AI clean decision
- package.json names the CLI bin whalent as dist/index.cjs and its scripts are build, prepublishOnly, start, and test, with no preinstall, install, or postinstall hook.
- The published runtime identifies its service host as memory.whalent.com, matching the Whalent Memory CLI described by the package.
- optionalDependencies pins the sibling package @whalent/agent-core to the same version 0.3.459 and does not depend on this package name.
Evidence against
- Shipped bundles are obfuscated and contain child_process use plus writers for Claude MCP config, which execute only after the user launches the whalent command.
- new Function appears inside bundled schema code generation rather than an install-time payload.
Behavioral surface
CryptoDynamicRequireEnvironmentVarsEvalFilesystemNetwork
HighEntropyStringsMinifiedObfuscated
Source & flagged code
7 flagged · loading sourcedist/mcp.cjsView file
1#!/usr/bin/env node
L2: 'use strict';var a0_0xc47212=a0_0x4ba4;(function(_0x5c6320,_0x26c665){var _0x1567a0=a0_0x4ba4,_0x5678b8=_0x5c6320();while(!![]){try{var _0x41d91d=parseInt(_0x1567a0(0x21ae))/0x1+pa...
L3: || (${_0x46a83f} == "string" && ${_0x55d15b} && ${_0x55d15b} == +${_0x55d15b})`)[_0x3d794f(0x47f)](_0x22eac4,(0x0,_0x5d5716['_'])`+${_0x55d15b}`);return;case'integer':_0x127c7c[_0x...
High
Obfuscated Payload Loader
Source contains an obfuscated payload loader that reconstructs and executes hidden code.
dist/mcp.cjsView on unpkg · L1•stage = ast_semantic_analysis; reason = ast_path_work_budget_exceeded; limitedFiles = 1
High
Semantic Analysis Limited
A bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.
dist/mcp.cjsView on unpkg4|| (${_0x46a83f} === "string" && ${_0x55d15b} && ${_0x55d15b} == +${_0x55d15b} && !(${_0x55d15b} % 1))`)[_0x3d794f(0x47f)](_0x22eac4,(0x0,_0x5d5716['_'])`+${_0x55d15b}`);return;cas...
L5: || ${_0x46a83f} === "boolean" || ${_0x55d15b} === null`)['assign'](_0x22eac4,(0x0,_0x5d5716['_'])`[${_0x55d15b}]`);}}}function _0x14a963({gen:_0x2006c6,parentData:_0xc25b58,parentD...
L6: missingProperty: ${_0x83eb34},
Medium
Findings
3 High7 Medium7 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/core.cjs
MediumOversized Cli Entrypointdist/core.cjs
MediumStructural Risk Force Deep Review
LowNon Install Lifecycle Scripts
LowScripts Present
LowEvaldist/mcp.cjs
LowFilesystem
LowHigh Entropy Strings
Low
Affected versions and remediation
This report applies to @whalent/agent@0.3.459.
See all versions of @whalent/agent for other recorded verdicts.
- Review the evidence and your use of @whalent/agent@0.3.459 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.