registry / @whalent/agent / 0.3.523
@whalent/agent@0.3.523
⚠ Under reviewStable wrapper for Whalent Agent core runtime
AI Security Review
scanned 4d ago · by lpm-firewall-aiThe manifest indicates a CLI package without an npm install hook, while the obfuscated MCP bundle contains skill download and installation behavior. The available source did not resolve how that behavior is activated or whether consent gates it, so manual review is appropriate.
Static reason
One or more suspicious static signals were detected.
Trigger
Unknown; inspection did not establish what command or event invokes the skill installation helper.
Impact
If invoked without suitable consent, downloaded skill content could change agent behavior; the source did not establish whether that condition occurs.
Mechanism
The helper fetches skill data from a supplied service base and writes files into an agent, workspace, or machine skill directory, with a path containment check.
AI rationale
The manifest indicates a CLI package without an npm install hook, while the obfuscated MCP bundle contains skill download and installation behavior. The available source did not resolve how that behavior is activated or whether consent gates it, so manual review is appropriate.
Evidence
dist/mcp.cjs
Decision evidence
public snapshotAI called this Manual Review at 78.0% confidence as Unknown with medium false-positive risk.
AI review observations
- The manifest exposes a CLI entrypoint and declares a prepublishOnly build hook, with no npm install hook.
- The indicated MCP bundle location maps agent platforms to their configuration directories; nearby code resolves skill destinations for agent, workspace, and machine scopes.
- The MCP bundle downloads skill data from a supplied service base and writes files after checking that their paths stay within the selected destination.
Evidence against
Behavioral surface
CryptoDynamicRequireEnvironmentVarsEvalFilesystemNetwork
HighEntropyStringsMinifiedObfuscatedUrlStrings
NoLicense
Oversized source lightweight scan
Source & flagged code
7 flagged · loading sourcedist/mcp.cjsView file
1#!/usr/bin/env node
L2: 'use strict';var a0_0x431fa1=a0_0x2155;(function(_0x2c3d56,_0x55841e){var _0x533107=a0_0x2155,_0x2e4c60=_0x2c3d56();while(!![]){try{var _0x495440=-parseInt(_0x533107(0x188a))/0x1+-...
L3: || (${_0x367679} == "string" && ${_0x2ce0f9} && ${_0x2ce0f9} == +${_0x2ce0f9})`)[_0x3b3ee1(0x223c)](_0x24a705,(0x0,_0x36af04['_'])`+${_0x2ce0f9}`);return;case _0x3b3ee1(0x207):_0x4...
High
Obfuscated Payload Loader
Source contains an obfuscated payload loader that reconstructs and executes hidden code.
dist/mcp.cjsView on unpkg · L1•stage = ast_semantic_analysis; reason = ast_path_work_budget_exceeded; limitedFiles = 1
High
Semantic Analysis Limited
A bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.
dist/mcp.cjsView on unpkg4|| (${_0x367679} === "string" && ${_0x2ce0f9} && ${_0x2ce0f9} == +${_0x2ce0f9} && !(${_0x2ce0f9} % 1))`)['assign'](_0x24a705,(0x0,_0x36af04['_'])`+${_0x2ce0f9}`);return;case'boolea...
L5: || ${_0x367679} === "boolean" || ${_0x2ce0f9} === null`)['assign'](_0x24a705,(0x0,_0x36af04['_'])`[${_0x2ce0f9}]`);}}}function _0x11fb67({gen:_0x1a3630,parentData:_0x279290,parentD...
L6: missingProperty: ${_0x50c9db},
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@0.3.523.
See all versions of @whalent/agent for other recorded verdicts.
- Review the evidence and your use of @whalent/agent@0.3.523 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.