@funny-booth/agent-core@0.1.19
古い版です。最新版を使ってください
AIエージェント起動CLI（共有ランチャー + MCPサーバー）
AI Security Review
scanned 12d ago · by lpm-firewall-aiThe portal-driven command execution presents a concrete but caller-invoked remote code execution risk, so warn rather than block. The feedback guidance is also concerning, but the inspected feedback path does not establish a hidden unrelated recipient.
Static reason
One or more suspicious static signals were detected.; previous stored version diff introduced dangerous source
Trigger
A caller invokes the MCP installation operation and the portal returns an install command.
Impact
The portal-provided command can run with the CLI user's privileges.
Mechanism
The package sends an authenticated request to the configured portal and passes the returned install command to child_process.
AI rationale
The portal-driven command execution presents a concrete but caller-invoked remote code execution risk, so warn rather than block. The feedback guidance is also concerning, but the inspected feedback path does not establish a hidden unrelated recipient.
Evidence
dist/index.js
Decision evidence
public snapshotAI called this Suspicious at 95.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- The published CLI entrypoint is dist/index.js.
- A portal response supplies an install command that the package executes through child_process.
- Feedback guidance tells an agent to send feedback without the user's request.
- Feedback posts to the configured portal URL with a bearer token.
Evidence against
- The command execution is inside the MCP install operation, rather than an observed package install or import hook.
- The feedback sender transmits caller-provided feedback data to the configured portal.
Behavioral surface
ChildProcessCryptoDynamicRequireEnvironmentVarsFilesystemNetwork
HighEntropyStringsMinifiedObfuscatedUrlStrings
NoLicense
Source & flagged code
7 flagged · loading sourcedist/index.jsView file
•matchType = previous_version_dangerous_delta
matchedPackage = @funny-booth/agent-core@0.1.18
matchedIdentity = npm:QGZ1bm55LWJvb3RoL2FnZW50LWNvcmU:0.1.18
similarity = 0.813
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/index.jsView on unpkg1#!/usr/bin/env node
L2: 'use strict';const _0x228fc0=_0x2529;(function(_0x3e9753,_0x457bc6){const _0x3e220b=_0x2529,_0xec1f82=_0x3e9753();while(!![]){try{const _0x2754cb=parseInt(_0x3e220b(0x1f5))/0x1+-pa...
High
1#!/usr/bin/env node
L2: 'use strict';const _0x228fc0=_0x2529;(function(_0x3e9753,_0x457bc6){const _0x3e220b=_0x2529,_0xec1f82=_0x3e9753();while(!![]){try{const _0x2754cb=parseInt(_0x3e220b(0x1f5))/0x1+-pa...
High
Same File Env Network Execution
A single source file combines environment access, network access, and code or shell execution; review context before blocking.
Findings
1 Critical6 High4 Medium6 Low
CriticalPrevious Version Dangerous Deltadist/index.js
HighChild Processdist/index.js
HighSame File Env Network Executiondist/index.js
HighCommand Output Exfiltrationdist/index.js#virtual:normalized:round1
HighObfuscated Payload Loaderdist/feedback.js
HighTrigger Reachable Command Output Exfiltrationdist/index.js#virtual:normalized:round1
HighObfuscated
MediumDynamic Requiredist/feedback.js
MediumNetwork
MediumEnvironment Vars
MediumStructural Risk Force Deep Review
LowNon Install Lifecycle Scripts
LowScripts Present
Low
Affected versions and remediation
This report applies to @funny-booth/agent-core@0.1.19.
See all versions of @funny-booth/agent-core for other recorded verdicts.
- Review the evidence and your use of @funny-booth/agent-core@0.1.19 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.