// npm 패키지

ciphernest

AI-aware security scanner for VS Code — code vulnerabilities, LLM risks, secrets, dependencies, MCP & agent security

npmjs.com tarball repo homepage

버전

메인테이너

최초 publish

2026-06-04

publisher

vmmuthu31

tarball

25,503,223 B

AUTO-PUBLISHED·2개 버전 인덱싱됨·최근 publish: 2026-06-05

// exfil path

what is read → where it ships

steals

● GitHub PAT
● AWS keys
● MCP config
● Chromium logins
● AI API keys
● Azure creds
○ clipboard
○ fs recursive read
○ home dir

sends to

(no destination string extracted — payload may be dynamic / obfuscated)

evidence in excerpt

> "url": "git+https://github.com/vmmuthu31/CipherNest.git"
> eval(code); // CODE001 + LLM003

→ view full payload

// offending code· @0.8.4· 3 files flagged

llm: malicious · 0.95

→ 크리덴셜 읽기 (reads-github-tokens, reads-aws-creds, reads-ai-api-keys, reads-azure-creds) + 외부 전송지 webhook-bin 조합 — 전형적인 유출 패턴.

@0.8.4·2026-06-05·AUTO-PUBLISHED·publisher: vmmuthu31
heuristic 55/100
static flags 15
llm malicious (0.95) via fast-track
tarball-size-jump:4xnew-publisher:1dhas-source-repopublic-github-pusheval-dynamicreads-env-varsclipboard-accessreads-github-tokenschild-process-spawnfs-recursive-readreads-aws-credsreads-mcp-configreads-chromium-credsreads-ai-api-keysreads-homedirwebhook-binbase64-decodereads-azure-creds

--- package/package.json (excerpt) --- { "name": "ciphernest", "displayName": "CipherNest", "description": "AI-aware security scanner for VS Code — code vulnerabilities, LLM risks, secrets, dependencies, MCP & agent security", "version": "0.8.4", "publisher": "CipherNestSecure", "icon": "media/logo.png", "repository": { "type": "git", "url": "git+https://github.com/vmmuthu31/CipherNest.git" }, "engines": { "vscode": "^1.85.0" }, "categories": [ "Linters", "Other" ], "keywords": [ "security", "llm", "ai", "secrets", "vulnerabilities", "mcp", "agent" ], "activationEvents": [ "onStartupFinished" ], "main": "./out/extension.js", "contributes": { "commands": [ { "command": "ciphernest.scanFile", "title": "CipherNest: Scan Current File", "icon": "$(shield)" }, { "command": "ciphernest.scanWorkspace", "title": "CipherNest: Scan Workspace", "icon": "$(search)" }, { "command": "ciphernest.scanDependencies", "title": "CipherNest: Scan Dependencies", "icon": "$(package)" }, { "command": "ciphernest.simulatePromptAttack", "title": "CipherNest: Simulate Prompt Attack", "icon": "$(bug)" }, { "command": "ciphernest.showDashboard", "title": "CipherNest: Show Security Dashboard", "icon": "$(graph)" }, { "command": "ciphernest.showAttackGraph", --- package/test-fixtures/vulnerable-sample.ts (excerpt) --- // CipherNest test fixture — intentionally vulnerable code // DO NOT use in production import OpenAI from 'openai'; import { AgentExecutor } from 'langchain/agents'; const openai = new OpenAI({ apiKey: 'sk-proj-abc123XYZ789aaabbbcccdddeee' }); // SEC001 + CODE008 // LLM001: user input passed directly to LLM async function handleChat(userInput: string) { const response = await openai.chat.completions.create({ model: 'gpt-4', messages: [{ role: 'user', content: userInput }], // LLM001 }); // LLM003: executing LLM output const code = response.choices[0].message.content!; eval(code); // CODE001 + LLM003 // CODE004: XSS document.getElementById('output')!.innerHTML = code; // CODE004 return code; } // CODE009: TLS disabled const tlsAgent = require('https').globalAgent; const opts = { rejectUnauthorized: false }; // CODE009 // CODE010: weak crypto import { createHash } from 'crypto'; const hash = createHash('md5').update('password').digest('hex'); // CODE010 // LLM006: agent with shell tool const executor = new AgentExecutor({ tools: [new ShellTool(), new SQLDatabase()], // LLM006, LLM008 maxIterations: undefined, // AGENT003 }); // LLM011: agent without restrictions const agent = initialize_agent(tools, llm); // LLM011 // CODE015: CORS wildcard res.setHeader('Access-Control-Allow-Origin', '*'); // CODE015 --- package/src/extension.ts (excerpt) --- import * as vscode from 'vscode'; import { SecurityEngine } from './scanner/index'; import { SecurityEngine as CoreEngine } from '@ciphernest/core'; import { scanDependencies } from './scanner/dependencyScanner'; import { scanInstalledExtensions } from './scanner/extensionScanner'; import { DiagnosticsManager } from './ui/diagnostics'; import { FindingsTreeProvider } from './ui/findingsTreeProvider'; import { SecurityCodeLensProvider } from './ui/securityCodeLens'; import { SecurityFileDecorationProvider } from './ui/fileDecorations'; import { DashboardPanel } from './ui/dashboardPanel'; import { PromptAttackPanel } from './ui/promptAttackPanel'; import { NpmShieldPanel } from './ui/panels/npmShieldPanel'; import { AgentPermissionMapPanel } from './ui/panels/agentPermissionMapPanel'; import { SecretJourneyPanel } from './ui/panels/secretJourneyPanel'; import { ShipCheckPanel } from './ui/panels/shipCheckPanel'; import { BackgroundScanner } from './daemon/backgroundScanner'; import { FindingTriageStore } from './triage/suppressionStore'; import { AuditLogger } from './audit/auditLog'; import { PolicyEngine } from './policy/policyEngine'; import { SecurityMemoryStore } from './memory/securityMemory'; import { AutoFixProvider } from './autofix/autoFixEngine'; import { AttackSurfaceMapPanel } from './ui/panels/attackSurfaceMapPanel'; import { ScoreViewProvider } from './ui/scoreViewProvider'; import { writeSbomFile, generateSbom } from './sbom/sbomGenerator'; import { Finding } f

--- package/test-fixtures/vulnerable-sample.ts (excerpt) --- // CipherNest test fixture — intentionally vulnerable code // DO NOT use in production import OpenAI from 'openai'; import { AgentExecutor } from 'langchain/agents'; const openai = new OpenAI({ apiKey: 'sk-proj-abc123XYZ789aaabbbcccdddeee' }); // SEC001 + CODE008 // LLM001: user input passed directly to LLM async function handleChat(userInput: string) { const response = await openai.chat.completions.create({ model: 'gpt-4', messages: [{ role: 'user', content: userInput }], // LLM001 }); // LLM003: executing LLM output const code = response.choices[0].message.content!; eval(code); // CODE001 + LLM003 // CODE004: XSS document.getElementById('output')!.innerHTML = code; // CODE004 return code; } // CODE009: TLS disabled const tlsAgent = require('https').globalAgent; const opts = { rejectUnauthorized: false }; // CODE009 // CODE010: weak crypto import { createHash } from 'crypto'; const hash = createHash('md5').update('password').digest('hex'); // CODE010 // LLM006: agent with shell tool const executor = new AgentExecutor({ tools: [new ShellTool(), new SQLDatabase()], // LLM006, LLM008 maxIterations: undefined, // AGENT003 }); // LLM011: agent without restrictions const agent = initialize_agent(tools, llm); // LLM011 // CODE015: CORS wildcard res.setHeader('Access-Control-Allow-Origin', '*'); // CODE015 --- package/src/extension.ts (excerpt) --- import * as vscode from 'vscode'; import { SecurityEngine } from './scanner/index'; import { scanDependencies } from './scanner/dependencyScanner'; import { scanInstalledExtensions } from './scanner/extensionScanner'; import { DiagnosticsManager } from './ui/diagnostics'; import { FindingsTreeProvider } from './ui/findingsTreeProvider'; import { DashboardPanel } from './ui/dashboardPanel'; import { PromptAttackPanel } from './ui/promptAttackPanel'; import { NpmShieldPanel } from './ui/panels/npmShieldPanel'; import { AgentPermissionMapPanel } from './ui/panels/agentPermissionMapPanel'; import { SecretJourneyPanel } from './ui/panels/secretJourneyPanel'; import { ShipCheckPanel } from './ui/panels/shipCheckPanel'; import { BackgroundScanner } from './daemon/backgroundScanner'; import { FindingTriageStore } from './triage/suppressionStore'; import { AuditLogger } from './audit/auditLog'; import { PolicyEngine } from './policy/policyEngine'; import { SecurityMemoryStore } from './memory/securityMemory'; import { AutoFixProvider } from './autofix/autoFixEngine'; import { AttackSurfaceMapPanel } from './ui/panels/attackSurfaceMapPanel'; import { scanCicdFiles } from './scanner/cicdScanner'; import { scanDockerFiles } from './scanner/dockerScanner'; import { scanIaCFiles } from './scanner/iacScanner'; import { writeSbomFile, generateSbom } from './sbom/sbomGenerator'; import { Finding } from './types'; import { computeScore } from './scoring/scoreEngine'; let statusBarItem: vscode --- package/src/scanner/cicdScanner.ts (excerpt) --- import * as path from 'path'; import * as fs from 'fs'; import { Finding } from '../types'; interface CicdRule { id: string; name: string; description: string; severity: Finding['severity']; pattern: RegExp; remediation: string; platform: 'github' | 'gitlab' | 'any'; } const CICD_RULES: CicdRule[] = [ // GitHub Actions — high-severity { id: 'CICD001', platform: 'github', severity: 'critical', name: 'pull_request_target with code checkout', description: 'pull_request_target gives workflows write permissions and access to secrets — combined with checkout of PR code this enables arbitrary code execution', pattern: /pull_request_target/g, remediation: 'Replace with pull_request trigger. If you must use pull_request_target, never checkout or run untrusted PR code in the same job.', }, { id: 'CICD002', platform: 'github', severity: 'critical', name: 'Unpinned action (uses: @latest or branch ref)', description: 'Using action refs like @main, @master, or @latest is vulnerable to supply chain attacks — the action can be silently changed', pattern: /uses:\s+[a-zA-Z0-9_/-]+@(?:main|master|latest|v\d+(?!\.\d))\b/g, remediation: 'Pin all actions to a specific commit SHA: uses: actions/checkout@a5ac7e51b41094c92402da3b24376905380afc29', }, { id: 'CICD003', platform: 'github', severity: 'high', name: 'Script injection via github.event context', description: 'Using ${{ github.event.* }} directly in run: steps ena

{ "name": "ciphernest", "displayName": "CipherNest", "description": "AI-aware security scanner for VS Code — code vulnerabilities, LLM risks, secrets, dependencies, MCP & agent security", "version": "0.8.4", "publisher": "CipherNestSecure", "icon": "media/logo.png", "repository": { "type": "git", "url": "git+https://github.com/vmmuthu31/CipherNest.git" }, "engines": { "vscode": "^1.85.0" }, "categories": [ "Linters", "Other" ], "keywords": [ "security", "llm", "ai", "secrets", "vulnerabilities", "mcp", "agent" ], "activationEvents": [ "onStartupFinished" ], "main": "./out/extension.js", "contributes": { "commands": [ { "command": "ciphernest.scanFile", "title": "CipherNest: Scan Current File", "icon": "$(shield)" }, { "command": "ciphernest.scanWorkspace", "title": "CipherNest: Scan Workspace", "icon": "$(search)" }, { "command": "ciphernest.scanDependencies", "title": "CipherNest: Scan Dependencies", "icon": "$(package)" }, { "command": "ciphernest.simulatePromptAttack", "title": "CipherNest: Simulate Prompt Attack", "icon": "$(bug)" }, { "command": "ciphernest.showDashboard", "title": "CipherNest: Show Security Dashboard", "icon": "$(graph)" }, { "command": "ciphernest.showAttackGraph",

// CipherNest test fixture — intentionally vulnerable code // DO NOT use in production import OpenAI from 'openai'; import { AgentExecutor } from 'langchain/agents'; const openai = new OpenAI({ apiKey: 'sk-proj-abc123XYZ789aaabbbcccdddeee' }); // SEC001 + CODE008 // LLM001: user input passed directly to LLM async function handleChat(userInput: string) { const response = await openai.chat.completions.create({ model: 'gpt-4', messages: [{ role: 'user', content: userInput }], // LLM001 }); // LLM003: executing LLM output const code = response.choices[0].message.content!; eval(code); // CODE001 + LLM003 // CODE004: XSS document.getElementById('output')!.innerHTML = code; // CODE004 return code; } // CODE009: TLS disabled const tlsAgent = require('https').globalAgent; const opts = { rejectUnauthorized: false }; // CODE009 // CODE010: weak crypto import { createHash } from 'crypto'; const hash = createHash('md5').update('password').digest('hex'); // CODE010 // LLM006: agent with shell tool const executor = new AgentExecutor({ tools: [new ShellTool(), new SQLDatabase()], // LLM006, LLM008 maxIterations: undefined, // AGENT003 }); // LLM011: agent without restrictions const agent = initialize_agent(tools, llm); // LLM011 // CODE015: CORS wildcard res.setHeader('Access-Control-Allow-Origin', '*'); // CODE015

import * as vscode from 'vscode'; import { SecurityEngine } from './scanner/index'; import { SecurityEngine as CoreEngine } from '@ciphernest/core'; import { scanDependencies } from './scanner/dependencyScanner'; import { scanInstalledExtensions } from './scanner/extensionScanner'; import { DiagnosticsManager } from './ui/diagnostics'; import { FindingsTreeProvider } from './ui/findingsTreeProvider'; import { SecurityCodeLensProvider } from './ui/securityCodeLens'; import { SecurityFileDecorationProvider } from './ui/fileDecorations'; import { DashboardPanel } from './ui/dashboardPanel'; import { PromptAttackPanel } from './ui/promptAttackPanel'; import { NpmShieldPanel } from './ui/panels/npmShieldPanel'; import { AgentPermissionMapPanel } from './ui/panels/agentPermissionMapPanel'; import { SecretJourneyPanel } from './ui/panels/secretJourneyPanel'; import { ShipCheckPanel } from './ui/panels/shipCheckPanel'; import { BackgroundScanner } from './daemon/backgroundScanner'; import { FindingTriageStore } from './triage/suppressionStore'; import { AuditLogger } from './audit/auditLog'; import { PolicyEngine } from './policy/policyEngine'; import { SecurityMemoryStore } from './memory/securityMemory'; import { AutoFixProvider } from './autofix/autoFixEngine'; import { AttackSurfaceMapPanel } from './ui/panels/attackSurfaceMapPanel'; import { ScoreViewProvider } from './ui/scoreViewProvider'; import { writeSbomFile, generateSbom } from './sbom/sbomGenerator'; import { Finding } f