registry / @whalent/agent / 0.3.519
@whalent/agent@0.3.519
Stable wrapper for Whalent Agent core runtime
AI Security Review
scanned 5d ago · by lpm-firewall-aiThe manifest shows no automatic install lifecycle hook, but the obfuscated bundle contains agent skill download and installation behavior whose caller and consent boundary remain unresolved. This supports a warning for agent extension lifecycle risk, not a confirmed malicious finding.
Static reason
One or more suspicious static signals were detected.
Trigger
Unresolved; the inspected excerpts do not show what invokes skill installation.
Impact
If reached without user authorization, this could add files that an agent may load; that activation and effect are not established by the inspected source.
Mechanism
The installer resolves a destination under an agent directory, creates it, and writes supplied files; another function fetches skill files from a configured service.
AI rationale
The manifest shows no automatic install lifecycle hook, but the obfuscated bundle contains agent skill download and installation behavior whose caller and consent boundary remain unresolved. This supports a warning for agent extension lifecycle risk, not a confirmed malicious finding.
Evidence
dist/mcp.cjs
Decision evidence
public snapshotAI called this Suspicious at 73.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- The package exposes an obfuscated bundled command-line entrypoint.
- The bundle resolves skill destinations under agent directories, including the home directory.
- A skill installer creates the destination and writes supplied files there.
- Skill files can be fetched from a configured service using an authorization header.
Evidence against
- The manifest has no install, preinstall, or postinstall hook; its only publication hook runs a build.
- The inspected code does not establish whether skill installation requires a user command or consent.
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_0x5c1c30=a0_0xfddb;(function(_0x9e5710,_0x4a9463){var _0x309782=a0_0xfddb,_0x5e7455=_0x9e5710();while(!![]){try{var _0x2acc59=parseInt(_0x309782(0x1804))/0x1+pa...
L3: || (${_0x590633} == "string" && ${_0xab5ece} && ${_0xab5ece} == +${_0xab5ece})`)['assign'](_0x437c83,(0x0,_0x33aeef['_'])`+${_0xab5ece}`);return;case'integer':_0x57afd7['elseIf']((...
High
Obfuscated Payload Loader
Source contains an obfuscated payload loader that reconstructs and executes hidden code.
dist/mcp.cjsView on unpkg · L1•stage = ast_semantic_analysis; reason = ast_path_work_budget_exceeded; limitedFiles = 1
High
Semantic Analysis Limited
A bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.
dist/mcp.cjsView on unpkg4|| (${_0x590633} === "string" && ${_0xab5ece} && ${_0xab5ece} == +${_0xab5ece} && !(${_0xab5ece} % 1))`)[_0x13b3c3(0x8f9)](_0x437c83,(0x0,_0x33aeef['_'])`+${_0xab5ece}`);return;cas...
L5: || ${_0x590633} === "boolean" || ${_0xab5ece} === null`)[_0x13b3c3(0x8f9)](_0x437c83,(0x0,_0x33aeef['_'])`[${_0xab5ece}]`);}}}function _0x431232({gen:_0x2f308e,parentData:_0x86565d...
L6: missingProperty: ${_0x27fe97},
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@0.3.519.
See all versions of @whalent/agent for other recorded verdicts.
- Review the evidence and your use of @whalent/agent@0.3.519 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.