registry / @whalent/agent / 0.3.508
@whalent/agent@0.3.508
Stable wrapper for Whalent Agent core runtime
AI Security Review
scanned 9d ago · by lpm-firewall-aiThe skill-management feature is activated through an MCP tool request, and the manifest has no npm install hook. Because it can add remote skill content to machine or workspace agent directories, this warrants a warning as a first-party agent extension capability.
Static reason
One or more suspicious static signals were detected.
Trigger
An MCP client invokes the skill installation action.
Impact
Installed skill content may affect later agent behavior; the inspected source does not establish covert execution or a confirmed attack.
Mechanism
The tool fetches the selected skill from the configured platform and writes its files into the resolved agent skill directory.
AI rationale
The skill-management feature is activated through an MCP tool request, and the manifest has no npm install hook. Because it can add remote skill content to machine or workspace agent directories, this warrants a warning as a first-party agent extension capability.
Evidence
dist/mcp.cjs
Decision evidence
public snapshotAI called this Suspicious at 82.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- The flagged MCP location resolves agent-specific directories; nearby code supports machine and workspace skill locations.
- An incoming MCP tool call can request a skill download and installation using a selected slug, platform URL, and bearer token.
- The installer writes returned skill files beneath the resolved directory and rejects paths outside that directory.
- The manifest's prepublishOnly hook runs the build; no install hook is declared.
Evidence against
- The skill installation is activated through an MCP tool request, not an npm lifecycle hook.
- The source shows a first-party skill-management feature, with no evidence that the downloaded content is automatically executed.
Behavioral surface
CryptoDynamicRequireEnvironmentVarsEvalFilesystemNetwork
HighEntropyStringsMinifiedObfuscatedUrlStrings
Source & flagged code
7 flagged · loading sourcedist/mcp.cjsView file
1#!/usr/bin/env node
L2: 'use strict';var a0_0x37c31f=a0_0x9005;(function(_0x340e88,_0x1757cf){var _0x1b8492=a0_0x9005,_0x240835=_0x340e88();while(!![]){try{var _0x3d3491=parseInt(_0x1b8492(0x1a2d))/0x1+pa...
L3: || (${_0x5a786b} == "string" && ${_0x56199d} && ${_0x56199d} == +${_0x56199d})`)[_0x2033c5(0x761)](_0x46ac02,(0x0,_0x24a27a['_'])`+${_0x56199d}`);return;case'integer':_0xe3b01b[_0x...
High
Obfuscated Payload Loader
Source contains an obfuscated payload loader that reconstructs and executes hidden code.
dist/mcp.cjsView on unpkg · L1•stage = ast_semantic_analysis; reason = ast_path_work_budget_exceeded; limitedFiles = 1
High
Semantic Analysis Limited
A bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.
dist/mcp.cjsView on unpkg4|| (${_0x5a786b} === "string" && ${_0x56199d} && ${_0x56199d} == +${_0x56199d} && !(${_0x56199d} % 1))`)['assign'](_0x46ac02,(0x0,_0x24a27a['_'])`+${_0x56199d}`);return;case'boolea...
L5: || ${_0x5a786b} === "boolean" || ${_0x56199d} === null`)['assign'](_0x46ac02,(0x0,_0x24a27a['_'])`[${_0x56199d}]`);}}}function _0x569239({gen:_0x5cdeb7,parentData:_0x2fa0c6,parentD...
L6: missingProperty: ${_0x41021b},
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@0.3.508.
See all versions of @whalent/agent for other recorded verdicts.
- Review the evidence and your use of @whalent/agent@0.3.508 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.