---
id: "GHSA-fwcp-q8x6-4fc2"
kind: "package"
aliases: ["GHSA-fwcp-q8x6-4fc2", "MAL-2026-17384"]
title: "Malicious code in @xayz/baileys (npm)"
severity: "critical"
cvss: null
epss: null
exploited: false
status: "open"
category: "package"
products: [{"name": "@xayz/baileys", "fixed": [], "affected": "all versions", "ecosystem": "npm"}]
ecosystems: ["npm"]
tags: ["malicious", "package"]
not_affected_if: []
published_at: "2026-10-08T13:31:42+00:00"
first_seen_at: "2026-10-29T14:24:36+00:00"
updated_at: "2026-10-29T14:25:47+00:00"
sources: [{"source": "osv-malicious", "url": "https://osv.dev/vulnerability/MAL-2026-17384", "first_seen_at": "2026-10-29T14:24:36+00:00"}]
---

# Malicious code in @xayz/baileys (npm)

## Affected

- npm:@xayz/baileys affected all versions fixed ?

## Timeline

- 2026-10-29T14:24:36+00:00 osv-malicious https://osv.dev/vulnerability/MAL-2026-17384

## From the sources

This package is part of a large family (100+ identified as of September 2026) of near-identical
forks of the Baileys WhatsApp Web library that inject a covert channel-subscription action into
the WhatsApp socket layer. On connect, the injected code issues an authenticated `w:mex` FOLLOW
query (query_id 7871414976211147) against one or more attacker-chosen WhatsApp Channel/Newsletter
JIDs, using the installer's own authenticated WhatsApp session -- silently subscribing the
victim's account to channels it never asked to join.

The target JID(s) are hidden from casual source review via one of several obfuscation techniques
observed across the family: a plain string literal, base64 encoding, base64+XOR, or a char-code
array reconstructed at runtime. Some variants instead fetch a mutable, attacker-controlled remote
JSON/JS list of target JIDs from GitHub or another host at runtime, letting the target list change
after installation without a new npm publish. Every sample in the family shares the same underlying
mechanism (a wrapped/patched socket-connect routine that fires the FOLLOW query some seconds after
connect), even though the package name, JID value(s), and obfuscation/delivery method differ per
fork.

The malicious action abuses the installer's own authenticated WhatsApp session to gain reach and
subscribers for attacker-controlled channels; it does not exfiltrate credentials, establish
persistence, or execute arbitrary remote code.

Affected package: @xayz/baileys (npm), version(s): 6.1.8, 6.1.7, 6.1.6, 6.1.5, 6.1.4, 6.1.3, 6.1.2, 6.1.1, 6.1.0, 6.0.75, 6.0.6, 6.0.5, 6.0.4, 6.0.3, 6.0.2, 6.0.1, 6.0.0, 5.0.0, 4.0.0, 3.0.0.

---
_-= Per source details. Do not edit below this line.=-_

## Source: amazon-inspector (3970c07a60e90f00f876e1f17c68e2eac1de32b845c00599ccf54b522b2d7245)
On the first successful WhatsApp connection via makeWASocket, the library issues a FOLLOW query for a hardcoded author-controlled newsletter JID (120363427430697245@newsletter) using the end user's authenticated WhatsApp session. The behavior is driven by a DEFAULT_AUTO_FOLLOW_CHANNELS constant consumed by the connection 'open' handler, which calls newsletterWMexQuery(jid, QueryIds.FOLLOW) for each entry. The behavior is documented and can be disabled via autoFollowChannels:false, but the default-on configuration uses the installer-user's WhatsApp identity to inflate the author's channel reach on a third-party platform. No credential theft, no exfiltration of installer secrets, no remote code execution, no persistence, and no install- or import-time destructive action; the ping-related static match in lib/Utils/generics.js corresponds to a WebSocket keepalive/latency utility, not network exfiltration.

## Source: ghsa-malware (b89023f96479d3074004c89716c5ed91b7bd86f978b92b69675c4d2e124d4f01)
Any computer that has this package installed or running should be considered fully compromised. All secrets and keys stored on that computer should be rotated immediately from a different computer. The package should be removed, but as full control of the computer may have been given to an outside entity, there is no guarantee that removing the package will remove all malicious software resulting from installing it.

---

_Free to read at a human pace. An API with keys and an MCP server are in preparation; early access: hello@infrastack.ai_