registry / @walkhi/code-relax / 0.1.0-beta.65

@walkhi/code-relax@0.1.0-beta.65

Code Relax Windows, macOS and Linux workstation service

AI Security Review

scanned 1d ago · by lpm-firewall-ai

The global-install postinstall path automatically enables login startup and places a package-owned skill in the shared agent skills directory. This fits agent extension lifecycle risk; the inspected behavior does not establish broader control-surface abuse or a concrete attack.

Static reason

High-risk behavior combination matched malicious policy.; source matched previously finalized malicious package; routed for review; previous stored version diff introduced dangerous source

Trigger

Installing the package globally with npm.

Impact

The Code Relax service can start at login, and its skill becomes available in the user’s agent environment without a separate setup command.

Mechanism

The postinstall hook runs setup, which enables resident startup and writes the package-owned skill under the user’s home directory.

AI rationale

Evidence

package.jsontools/postinstall.mjsdist/bin/codex-remote.mjsdist/src/skill-install.mjsdist/skills/code-relax/SKILL.mddist/src/platform/linux/resident-startup.mjs~/.agents/skills/code-relax/SKILL.md~/.config/autostart/cn.walkhi.code-relax.desktop

Decision evidence

public snapshot

AI called this Suspicious at 88.0% confidence as Dangerous Capability with medium false-positive risk.

Evidence for warning

The manifest runs a postinstall hook; it proceeds with setup only when invoked directly during a global npm install.
Setup enables resident startup and installs this package’s skill in the user’s shared agent skills directory.
On Linux, resident startup targets the user’s XDG autostart directory.
The inspected dots-room receiver parses ChatGPT SSE data frames and handles completion markers; it does not establish an account action.

Evidence against

The skill installer refuses to overwrite an existing skill unless it recognizes it as package-managed and unchanged.
The inspected runtime receiver behavior is event-stream parsing, not evidence of malicious activity.

Behavioral surface

Source

ChildProcessCryptoDynamicRequireEnvironmentVarsEvalFilesystemNetworkWebSocket

Supply chain

HighEntropyStringsMinified

Source & flagged code

19 flagged

package.jsonView file

•scripts.postinstall = node ./tools/postinstall.mjs

High

Install Time Lifecycle Scripts

Package defines install-time lifecycle scripts.

package.jsonView on unpkg

•scripts.postinstall = node ./tools/postinstall.mjs

Medium

Ambiguous Install Lifecycle Script

Install-time lifecycle script is not statically allowlisted and needs review.

package.jsonView on unpkg

dist/src/platform/windows/rustdesk-video-probe.mjsView file

77patternName = generic_password severity = medium line = 77

Medium

Secret Pattern

Package contains a possible secret pattern.

dist/src/platform/windows/rustdesk-video-probe.mjsView on unpkg · L77

tools/postinstall.mjsView file

1import { execFileSync } from 'node:child_process'; L2: import fs from 'node:fs';

High

Child Process

Package source references child process execution.

tools/postinstall.mjsView on unpkg · L1

8const pathsModule = new URL('../dist/src/platform/runtime-paths.mjs', import.meta.url); L9: const { runtimePaths } = await import(fs.existsSync(pathsModule) ? pathsModule : new URL('../src/platform/runtime-paths.mjs', import.meta.url)); L10: const { serviceRoot, stateRoot } = runtimePaths();

Medium

Dynamic Require

Package source references dynamic require/import behavior.

tools/postinstall.mjsView on unpkg · L8

dist/src/server.mjsView file

3import fs from 'node:fs'; L4: import http from 'node:http'; L5: import { createPowerManager } from './service-power.mjs'; ... L60: const messageReceipts = new messageDelivery.Store(); L61: const desktopHelper = process.platform === 'win32' ? new DesktopHelper() : null; L62: ... L85: while (true) { L86: const manifest = readRecord(path.join(directory, 'package.json')); L87: if (manifest?.name === '@walkhi/code-relax') return manifest.version || '未知'; ... L98: const webRoot = path.resolve(path.dirname(fileURLToPath(import.meta.url)), '..', 'web'); L99: const projectRoot = os.homedir(); L100: const { codexDataRoot, stateRoot, sessionsRoot, uploadsRoot } = runtimePaths();

High

Obfuscated Payload Loader

Source contains an obfuscated payload loader that reconstructs and executes hidden code.

dist/src/server.mjsView on unpkg · L3

Medium

Unsafe Vm Context

Package source executes code through a VM context API.

dist/src/server.mjsView on unpkg · L3

dist/bin/release-upload-server.mjsView file

4import path from 'node:path'; L5: import http from 'node:http'; L6: import { createHash, randomUUID } from 'node:crypto'; ... L8: import { Transform, Readable } from 'node:stream'; L9: import { execFile } from 'node:child_process'; L10: import { promisify } from 'node:util'; ... L19: const execute = promisify(execFile); L20: const config = JSON.parse(fs.readFileSync('/etc/codex-relay/android-releases.json', 'utf8')); L21: const credential = JSON.parse(fs.readFileSync('/etc/codex-relay/tencent-monitor.json', 'utf8')); ... L57: const server = http.createServer(async (req, res) => { L58: const reply = (code, data) => { if (!res.destroyed) { res.writeHead(code, { 'Content-Type': 'application/json', 'Cache-Control': 'no-store' }); res.end(JSON.stringify(data)); } }; L59: let claims;

Low

Weak Crypto

Package source references weak cryptographic algorithms.

dist/bin/release-upload-server.mjsView on unpkg · L4

dist/src/platform/macos/resident-startup.mjsView file

3import path from 'node:path'; L4: import { execFile } from 'node:child_process'; L5: import { promisify } from 'node:util'; ... L30: <key>EnvironmentVariables</key><dict>${Object.entries(variables).map(([key, value]) => `<key>${key}</key><string>${xml(value)}</string>`).join('')}</dict> L31: <key>StandardOutPath</key><string>${xml(path.join(stateRoot, 'login.stdout.log'))}</string> L32: <key>StandardErrorPath</key><string>${xml(path.join(stateRoot, 'login.stderr.log'))}</string> ... L40: if (uid === 0) throw new Error('请以普通 Mac 用户安装 Code Relax，不使用 sudo。'); L41: const home = os.homedir(); L42: const file = path.join(home, 'Library/LaunchAgents', `${launchAgentLabel}.plist`); ... L47: const { stdout } = await exec('/usr/bin/plutil', ['-convert', 'json', '-o', '-', file], { timeout: 5000

Medium

Install Persistence

Source writes installer persistence such as shell profile or service configuration.

dist/src/platform/macos/resident-startup.mjsView on unpkg · L3

dist/src/platform/windows/desktop-helper.mjsView file

1import net from 'node:net'; L2: import fs from 'node:fs'; L3: import path from 'node:path'; L4: import { execFile } from 'node:child_process'; L5: import { promisify } from 'node:util'; ... L10: const binary = fileURLToPath(new URL('../../../native/desktop-helper/CodeRelaxDesktop.exe', import.meta.url)); L11: const powershell = path.join(process.env.SystemRoot || 'C:/Windows', 'System32/WindowsPowerShell/v1.0/powershell.exe'); L12: const quote = value => `'${String(value).replaceAll("'", "''")}'`;

High

Same File Env Network Execution

A single source file combines environment access, network access, and code or shell execution; review context before blocking.

dist/src/platform/windows/desktop-helper.mjsView on unpkg · L1

dist/bin/codex-remote.mjsView file

26Install-time code directly mutates a foreign AI-agent control surface: L122: } else { L123: if (options.length) { fs.mkdirSync(stateRoot, { recursive: true }); writeRecord(powerConfigPath(stateRoot), { enabled: options[0] === 'on' }); } L124: result = { enabled: powerEnabled(stateRoot), active: false, error: '', serviceStopped: true }; ... L264: const source = path.join(stateRoot, name), target = path.join(physicalStateRoot, name); L265: if (fs.existsSync(source) && !fs.existsSync(target)) fs.copyFileSync(source, target, fs.constants.COPYFILE_EXCL); L266: } ... L377: installed.managedAppServer = bridge.managedAppServer; L378: installed.codexRuntime = runtime; L379: installed.managedAppServerMigrated = managedAppServerMigrated; Write operation from tools/postinstall.mjs: L26: function appen

Critical

Ai Agent Control Hijack

Source creates an unconsented AI-agent control surface through install-time mutation or a default unauthenticated remote skill channel.

dist/bin/codex-remote.mjsView on unpkg · L26

•matchType = previous_version_dangerous_delta matchedPackage = @walkhi/code-relax@0.1.0-beta.50 matchedIdentity = npm:QHdhbGtoaS9jb2RlLXJlbGF4:0.1.0-beta.50 similarity = 0.817 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/codex-remote.mjsView on unpkg

dist/src/skill-install.mjsView file

•Manifest-trigger-reachable source links an external AI-agent control path to a behavior-bearing write operation. import fs from 'node:fs'; import path from 'node:path'; import os from 'node:os'; import { createhash } from 'node:crypto'; import { readrecord, writerecord } from './service-lifecycle.mjs'; const owner = '@walkhi/code-relax'; const markername = '.code-relax-install.json'; const digest = value => createhash('sha256').update(value).digest('hex'); export const sk[redacted] = (home = os.homedir()) => path.join(home, '.agents/skills/code-relax'); export function assertplainpath(target) { for (let item = path.resolve(target); ; item = path.dirname(item)) { try { if (fs.lstatsync(item).issymboli

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.

dist/src/skill-install.mjsView on unpkg

dist/tools/rtc-route-helper.ps1View file

•path = dist/tools/rtc-route-helper.ps1 kind = build_helper sizeBytes = 5168 magicHex = [redacted]

Medium

Ships Build Helper

Package ships non-JavaScript build or shell helper files.

dist/tools/rtc-route-helper.ps1View on unpkg

dist/shared/code-browser.cjs#virtual:normalized:round1View file

•stage = ast_semantic_analysis; reason = ast_parse_error; limitedFiles = 4

High

Semantic Analysis Limited

A bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.

dist/shared/code-browser.cjs#virtual:normalized:round1View on unpkg

dist/src/background-process.mjsView file

•matchType = normalized_sha256 matchedPackage = @walkhi/code-relax@0.1.0-beta.1 matchedPath = dist/src/background-process.mjs matchedIdentity = npm:QHdhbGtoaS9jb2RlLXJlbGF4:0.1.0-beta.1 similarity = 1.000 summary = normalized source hash matched finalized malicious source

High

Known Malware Source Similarity

Source file is highly similar to a previously finalized malicious package; route for source-aware review.

dist/src/background-process.mjsView on unpkg

dist/src/platform/windows/network-info.mjsView file

•matchType = normalized_sha256 matchedPackage = @walkhi/code-relax@0.1.0-beta.1 matchedPath = [redacted]-info.mjs matchedIdentity = npm:QHdhbGtoaS9jb2RlLXJlbGF4:0.1.0-beta.1 similarity = 1.000 summary = normalized source hash matched finalized malicious source

High

Known Malware Source Similarity

Source file is highly similar to a previously finalized malicious package; route for source-aware review.

dist/src/platform/windows/network-info.mjsView on unpkg

dist/src/self-relay/windows-route-helper.mjsView file

•matchType = normalized_sha256 matchedPackage = @walkhi/code-relax@0.1.0-beta.1 matchedPath = dist/src/self-relay/windows-route-helper.mjs matchedIdentity = npm:QHdhbGtoaS9jb2RlLXJlbGF4:0.1.0-beta.1 similarity = 1.000 summary = normalized source hash matched finalized malicious source

High

Known Malware Source Similarity

Source file is highly similar to a previously finalized malicious package; route for source-aware review.

dist/src/self-relay/windows-route-helper.mjsView on unpkg

dist/web/vendor/markdown-it.umd.min.jsView file

•matchType = normalized_sha256 matchedPackage = @walkhi/code-relax@0.1.0-beta.1 matchedPath = dist/web/vendor/markdown-it.umd.min.js matchedIdentity = npm:QHdhbGtoaS9jb2RlLXJlbGF4:0.1.0-beta.1 similarity = 1.000 summary = normalized source hash matched finalized malicious source

High

Known Malware Source Similarity

Source file is highly similar to a previously finalized malicious package; route for source-aware review.

dist/web/vendor/markdown-it.umd.min.jsView on unpkg

Source & flagged code

19 flagged · loading source

package.jsonView file

•scripts.postinstall = node ./tools/postinstall.mjs

High

Install Time Lifecycle Scripts

Package defines install-time lifecycle scripts.

package.jsonView on unpkg

•scripts.postinstall = node ./tools/postinstall.mjs

Medium

Ambiguous Install Lifecycle Script

Install-time lifecycle script is not statically allowlisted and needs review.

package.jsonView on unpkg

dist/src/platform/windows/rustdesk-video-probe.mjsView file

77patternName = generic_password severity = medium line = 77

Medium

Secret Pattern

Package contains a possible secret pattern.

dist/src/platform/windows/rustdesk-video-probe.mjsView on unpkg · L77

Findings

2 Critical10 High9 Medium7 Low

CriticalAi Agent Control Hijackdist/bin/codex-remote.mjs

CriticalPrevious Version Dangerous Deltadist/bin/codex-remote.mjs

HighInstall Time Lifecycle Scriptspackage.json

HighChild Processtools/postinstall.mjs

HighSame File Env Network Executiondist/src/platform/windows/desktop-helper.mjs

HighObfuscated Payload Loaderdist/src/server.mjs

HighTrigger Reachable External Ai Agent Control Surface Mutationdist/src/skill-install.mjs

HighSemantic Analysis Limiteddist/shared/code-browser.cjs#virtual:normalized:round1

HighKnown Malware Source Similaritydist/src/background-process.mjs

HighKnown Malware Source Similaritydist/src/platform/windows/network-info.mjs

Affected versions and remediation

This report applies to @walkhi/code-relax@0.1.0-beta.65.

See all versions of @walkhi/code-relax for other recorded verdicts.

Review the evidence and your use of @walkhi/code-relax@0.1.0-beta.65 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: 2026-10-22 03:46:14.405Z (UTC).

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