modelence@0.28.0
The Node.js Framework for Real-Time MongoDB Apps
AI Security Review
scanned 8d ago · by lpm-firewall-aiThe CLI makes a project-scoped agent configuration change before obtaining affirmative consent for plugin installation, leaving a residual first-party extension setup risk. The reviewed credential transfers target the configured Modelence service, and the source does not support a malicious or publish-block verdict.
Static reason
One or more suspicious static signals were detected.; previous stored version diff introduced dangerous source
Trigger
A user runs `modelence setup`.
Impact
The project’s Claude Code configuration is changed to enable the plugin, creating a residual agent-extension setup risk.
Mechanism
Setup writes the Modelence marketplace and enables its plugin in `.claude/settings.json`; it does this before prompting, and without a terminal it does not prompt.
AI rationale
The CLI makes a project-scoped agent configuration change before obtaining affirmative consent for plugin installation, leaving a residual first-party extension setup risk. The reviewed credential transfers target the configured Modelence service, and the source does not support a malicious or publish-block verdict.
Evidence
dist/bin/modelence.js.claude/settings.json
Network endpoints1
github.com/modelence/modelence
Decision evidence
public snapshotAI called this Suspicious at 86.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- Running `modelence setup` calls code that registers and enables the project’s Modelence Claude plugin before a later optional installation prompt; the settings change also occurs without an interactive terminal.
- The manifest’s `prepare` script only configures the Git hooks path; it has no install, preinstall, or postinstall hook.
- The inspected setup and environment requests send Modelence tokens to the configured service endpoint. The CLI validates that endpoint and defaults to Modelence Cloud.
- The inspected channel joins are initiated by a client event and checked against channel access rules; the runtime query-cache callback handles removed queries.
Evidence against
- The Claude integration is the package’s own plugin and the setup command later presents an installation prompt, though the project setting is already changed before that prompt.
- The inspected source does not establish credential theft or another concrete attack beyond the project-scoped agent configuration change.
Behavioral surface
ChildProcessCrypto
Source & flagged code
9 flaggeddist/bin/modelence.jsView file
•matchType = previous_version_dangerous_delta
matchedPackage = modelence@0.27.1
matchedIdentity = npm:bW9kZWxlbmNl:0.27.1
similarity = 0.917
summary = stored previous version shares package body but lacks this dangerous source file
Critical
Previous Version Dangerous Delta
This package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
dist/bin/modelence.jsView on unpkg1#!/usr/bin/env node
L2: import {Command,Option}from'commander';import {readFileSync,promises,existsSync,createWriteStream,openAsBlob}from'fs';import {fileURLToPath}from'url';import ve,{join,dirname,relati...
L3: `,{mode:384});}async function Oe(e){let{hosts:t={}}=await B(),o=t[e];return !o?.token||!o.expiresAt||new Date(o.expiresAt).getTime()-eo<=Date.now()?null:o.token}async function _e(e...
High
Child Process
Package source references child process execution.
dist/bin/modelence.jsView on unpkg · L11#!/usr/bin/env node
L2: import {Command,Option}from'commander';import {readFileSync,promises,existsSync,createWriteStream,openAsBlob}from'fs';import {fileURLToPath}from'url';import ve,{join,dirname,relati...
L3: `,{mode:384});}async function Oe(e){let{hosts:t={}}=await B(),o=t[e];return !o?.token||!o.expiresAt||new Date(o.expiresAt).getTime()-eo<=Date.now()?null:o.token}async function _e(e...
Findings
1 Critical8 High4 Medium6 Low
CriticalPrevious Version Dangerous Deltadist/bin/modelence.js
HighChild Processdist/bin/modelence.js
HighShell
HighSame File Env Network Executiondist/bin/modelence.js
HighHost Fingerprint Exfiltrationdist/bin/modelence.js
HighCommand Output Exfiltrationdist/bin/modelence.js
HighSandbox Evasion Gated Capabilitydist/bin/modelence.js
HighTrigger Reachable Command Output Exfiltrationdist/bin/modelence.js
HighSemantic Analysis Limiteddist/chunk-4F3G5XRJ.js#virtual:normalized:round1
MediumDynamic Requiredist/chunk-2Z6CXEM6.js
MediumNetwork
MediumEnvironment Vars
Affected versions and remediation
This report applies to modelence@0.28.0.
See all versions of modelence for other recorded verdicts.
- Review the evidence and your use of modelence@0.28.0 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.