registry / Remote code execution / @raolin2025/claude-code-node / 2.8.21

@raolin2025/claude-code-node@2.8.21

Node.js AI Code Agent CLI - Zero dependencies, pure JavaScript, security hardened, multi-channel notifications, Telegram & QQ Bot remote programming, rich media upload, multi-account management

AI Security Review

scanned 2mo ago · by lpm-firewall-ai

No lifecycle hook, self-dependency, package-controlled exfiltration endpoint, or stealth install behavior was found. The explicit remote-agent design nevertheless includes an insufficiently guarded command-execution capability, so it should warn rather than block.

Static reason

High-risk behavior combination matched malicious policy.

Trigger

A configured Telegram or QQ bot receives a /run command while cc-notify is running.

Impact

An unauthorized chat participant could run commands as the local user and receive their output.

Mechanism

Remote chat input is passed to direct shell execution.

AI rationale

Evidence

package.jsonsrc/channel/notify-daemon.jssrc/channel/tg-listener.jssrc/core/config.jssrc/tools/npm-publish.js.claude-code/notify-api-key.txt~/.cc-node/cc-notify.pid~/.cc-node/cc-notify.log

Network endpoints3

api.telegram.orgapi.sgroup.qq.combots.qq.com/app/getAppAccessToken

Decision evidence

public snapshot

AI called this Suspicious at 90.0% confidence as Dangerous Capability with medium false-positive risk.

Evidence for warning

The cc-notify command executes raw /run message text with execSync and returns command output to chat.
Configured Telegram messages are forwarded directly into the common message handler.
Default QQ settings allow direct and group messages from every sender.
The publish tool reads an npm token, but uses it only for npm registry publishing.

Evidence against

package.json has no preinstall, install, or postinstall hook and no self-dependency.
The risky behavior requires running the cc-notify CLI and configuring a bot credential.
The local HTTP command path requires the configured API key.
No package-controlled exfiltration endpoint, stealth loader, or install-time mutation was found.

Behavioral surface

Source

ChildProcessCryptoEnvironmentVarsEvalFilesystemNetworkShellWebSocket

Supply chain

HighEntropyStrings

Source & flagged code

10 flagged

src/tools/bash.jsView file

11*/ L12: import { spawn } from 'child_process' L13: import { ToolDef } from '../types/index.js'

High

Child Process

Package source references child process execution.

src/tools/bash.jsView on unpkg · L11

8* - Windows (WSL) → wsl bash（bash 语义） L9: * - Windows (PS) → PowerShell L10: * - Windows (cmd) → cmd.exe 兜底

High

Shell

Package source references shell execution.

src/tools/bash.jsView on unpkg · L8

src/tools/npm-publish.jsView file

4* 封装了本项目 npm 发布的全部经验（见 NPM_STAGED_PUBLISH_GUIDE.md）： L5: * - 标准 `npm publish` 常因 npm CLI 强制 otplease/2FA 返回 403 L6: * - 最有效方法：bypass-2FA GAT token + 手动构造 PUT 请求 + `npm-auth-type: bearer`，无需 OTP ... L14: */ L15: import { execSync } from 'child_process' L16: import { readFileSync, writeFileSync, existsSync, mkdirSync } from 'fs' ... L25: 前置条件: L26: - 项目根目录（含 package.json） L27: - ~/.npmrc 中配置了 bypass-2FA 的 _authToken（npm_... 前缀） ... L97: export function parsePackJson(packJson) { L98: const data = JSON.parse(packJson) L99: const meta = Object.values(data)[0] || {}

Critical

Credential Exfiltration

Source appears to send environment or credential material to an external endpoint.

src/tools/npm-publish.jsView on unpkg · L4

Low

Weak Crypto

Package source references weak cryptographic algorithms.

src/tools/npm-publish.jsView on unpkg · L4

src/channel/notify-daemon.jsView file

30import { homedir } from 'node:os' L31: import { spawn, execSync } from 'node:child_process' L32: import { createConnection } from 'node:net' L33: import crypto from 'node:crypto' ... L42: // 生成或加载 API Key L43: let apiKey = process.env.CC_NOTIFY_API_KEY || '' L44: if (!apiKey) {

Critical

Same File Env Network Execution

A single source file combines environment access, network access, and code or shell execution with blocking evidence.

src/channel/notify-daemon.jsView on unpkg · L30

22Source sends the broad process environment to a literal external destination. L22: */ L23: import { createServer } from 'node:http' L24: import { ... L30: import { homedir } from 'node:os' L31: import { spawn, execSync } from 'node:child_process' L32: import { createConnection } from 'node:net' ... L42: // 生成或加载 API Key L43: let apiKey = process.env.CC_NOTIFY_API_KEY || '' L44: if (!apiKey) { L45: const configDir = join(process.cwd(), '.claude-code') L46: const configPath = join(configDir, 'notify-api-key.txt') ... L76: try {

Critical

Hardcoded Runtime Data Exfiltration

Source sends credentials or rich application records to a package-controlled external receiver enabled by default.

src/channel/notify-daemon.jsView on unpkg · L22

22Trigger-reachable chain: manifest.bin -> src/channel/notify-daemon.js L22: */ L23: import { createServer } from 'node:http' L24: import { ... L30: import { homedir } from 'node:os' L31: import { spawn, execSync } from 'node:child_process' L32: import { createConnection } from 'node:net' ... L42: // 生成或加载 API Key L43: let apiKey = process.env.CC_NOTIFY_API_KEY || '' L44: if (!apiKey) { L45: const configDir = join(process.cwd(), '.claude-code') L46: const configPath = join(configDir, 'notify-api-key.txt') ... L76: try {

src/security/ssrf-guard.jsView file

8*/ L9: import { lookup as dnsLookup } from 'dns' L10: import { isIP } from 'net'

High

Cloud Metadata Access

Source reaches cloud instance metadata or link-local credential endpoints.

src/security/ssrf-guard.jsView on unpkg · L8

src/security/enhanced-permission.jsView file

•Manifest-trigger-reachable source links an external AI-agent control path to a behavior-bearing write operation. ult.reasons.join('; ') : undefined, detail: result, } } case 'webfetch': { const url = input.url || '' if (!url) return { safe: true } const result = await checkurlsafety(url) return { safe: result.allowed, reason: result.reason, } } default: return { safe: true } } } _extractpattern(toolname, input) { switch (toolname) { case 'bash': return input.command || '' case 'read': case 'edit': case 'write': return input.file_path || '' case 'glob': case 'grep':

High

Trigger Reachable External Ai Agent Control Surface Mutation

Manifest-trigger-reachable source writes behavior-bearing configuration into a user or project AI-agent control surface.

src/security/enhanced-permission.jsView on unpkg

Source & flagged code

10 flagged · loading source

src/tools/bash.jsView file

11*/ L12: import { spawn } from 'child_process' L13: import { ToolDef } from '../types/index.js'

High

Child Process

Package source references child process execution.

src/tools/bash.jsView on unpkg · L11

8* - Windows (WSL) → wsl bash（bash 语义） L9: * - Windows (PS) → PowerShell L10: * - Windows (cmd) → cmd.exe 兜底

High

Shell

Package source references shell execution.

src/tools/bash.jsView on unpkg · L8

src/tools/npm-publish.jsView file

Findings

4 Critical5 High3 Medium6 Low

CriticalSame File Env Network Executionsrc/channel/notify-daemon.js

CriticalHardcoded Runtime Data Exfiltrationsrc/channel/notify-daemon.js

CriticalCredential Exfiltrationsrc/tools/npm-publish.js

CriticalTrigger Reachable Dangerous Capabilitysrc/channel/notify-daemon.js

HighChild Processsrc/tools/bash.js

HighShellsrc/tools/bash.js

HighCloud Metadata Accesssrc/security/ssrf-guard.js

HighTrigger Reachable External Ai Agent Control Surface Mutationsrc/security/enhanced-permission.js

HighTrigger Reachable Credential Exfiltrationsrc/channel/notify-daemon.js

MediumNetwork

MediumEnvironment Vars

Affected versions and remediation

This report applies to @raolin2025/claude-code-node@2.8.21.

See all versions of @raolin2025/claude-code-node for other recorded verdicts.

Review the evidence and your use of @raolin2025/claude-code-node@2.8.21 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: 2026-08-28 16:45:02.894Z (UTC).

Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Affected versions and remediation

Runtime surface

Behavioral surface

Source & flagged code