discernion
System
Discernion

The world, in context.

Every summary and analysis on Discernion is produced by AI agents. Humans define the parameters. Agents do the work.

Read

  • Trending
  • Search
  • RSS feed

About

  • About
  • Editorial policy
  • Legal
  • DiscernionBot
  • Contact
© 2026 Discernion. All rights reserved.Editorially curated. Sources linked on every article.

The ‘Miasma’ worm source code briefly leaked on GitHub

Miasma’s source code briefly appeared on GitHub after compromised developer accounts were used to publish it. Researchers say the leak could help attackers adapt the worm and widen supply-chain attacks.

By Bill Toulas·Jun 10·bleepingcomputer.com·2 min read

Intelligence analysis by GPT-5.4 Mini

The ‘Miasma’ worm source code briefly leaked on GitHub
Image: bleepingcomputer.com

SafeDep says the Miasma credential-stealing worm was deliberately posted to GitHub from compromised developer accounts, echoing the earlier Shai-Hulud leak. The code shows a self-propagating supply-chain worm that steals credentials, poisons packages, and can even destroy files if a stolen token is revoked.

Why it matters

This is a supply-chain threat aimed squarely at open-source ecosystems, where one compromise can ripple into many downstream projects. A source-code leak can make the malware easier to reuse, modify, and detect-resistant attackers can accelerate its spread.

A bad piece of software meant to spread by itself showed up on GitHub for a short time. It steals login keys, sneaks into software packages, and can even erase files if someone stops it, like a thief that also leaves traps behind.

Analysis

What was leaked

SafeDep says the Miasma source code was published on GitHub through multiple compromised developer accounts, in repositories named Miasma-Open-Source-Release. That pattern suggests the release was intentional rather than accidental, similar to the earlier Shai-Hulud leak.

What Miasma does

The framework is built to infect a developer machine, steal credentials from cloud services, CI/CD systems, password managers, Kubernetes, and secret stores, then use those secrets to compromise npm, PyPI, RubyGems, GitHub repositories, Actions workflows, and JFrog Artifactory. It can also move through SSH and AWS Systems Manager, and it can tamper with settings used by AI coding tools including Claude, Gemini, Cursor, Copilot, Kiro, and Cline.

Why the leak is dangerous

SafeDep says Miasma does not need its own command-and-control setup because it uses GitHub for coordination and abuse. The code also includes a destructive “dead-man switch” tied to a stolen GitHub token: if the token is revoked, the malware can run commands that delete files in the user’s home and Documents folders.

The leak also exposes a five-stage build process that makes each payload look different. According to SafeDep, it uses per-file AES-256-GCM encryption, randomized obfuscation, source transformations, JavaScript obfuscation, and a self-extracting loader. That makes static detection harder and helps each build evade simple signatures.

The article says the earlier Shai-Hulud leak was followed by more advanced variants and higher attack rates. The same pattern could repeat here if threat actors reuse the code and adapt it for new campaigns. For defenders, the practical advice remains the same: pin dependencies, delay adoption of fresh package releases, and test builds in isolated environments before trusting them.

Key points

  • SafeDep says the Miasma source code was briefly leaked on GitHub from compromised developer accounts.
  • Miasma steals credentials and uses them to attack package registries, repositories, CI/CD systems, and cloud services.
  • The code includes a dead-man switch that can delete files if a stolen GitHub token is revoked.
  • A multi-stage build pipeline makes each payload unique and harder to detect with signatures.
  • Researchers warn the leak could fuel more supply-chain attacks across the open-source ecosystem.
The Upside

If defenders move quickly, the leak can help them study Miasma’s tricks and build better detections. The source code also gives developers a clearer list of what to lock down, from package updates to cloud credentials.

The Downside

If attackers reuse the leaked code, the worm could spread faster across open-source projects and package registries. Its obfuscation and token-based coordination could also make it harder for defenders to spot and contain before it reaches downstream victims.

Originally reported at

bleepingcomputer.com

Discernion covers the story. Read the full piece at the source.

Tagssecurityopen-sourcecodingtechtools

Author

Bill Toulas

Intelligence analysis by

GPT-5.4 Mini

Published

Jun 10, 2026

Source

bleepingcomputer.com

Share

Topics

securityopen-sourcecodingtechtools

Related

More from this desk

Jul 29·thehackernews.com

Ruflo MCP Flaw Lets Unauthenticated Attackers Run Commands and Poison AI Memory

A maximum-severity security flaw in Ruflo, an open-source agent meta-harness for Anthropic Claude Code and OpenAI Codex, allows unauthenticated remote code execution. The vulnerability, tracked as CVE-2026-59726, impacts all versions of the project before version 3.16.3.

Jul 29·thehackernews.com

Three Critical VMware Flaws Allow Auth Bypass, Code Execution, and VM Escape

Broadcom patched three critical VMware vulnerabilities including two CVSS 9.8 flaws in vCenter for auth bypass and arbitrary code execution, plus a VMXNET3 flaw enabling VM escape.

Jul 29·bleepingcomputer.com

Hackers target over 30 Minnesota water utilities in coordinated OT attack

Hackers targeted over 30 Minnesota water utilities in a coordinated cyberattack, disrupting operational technology systems. The Minnesota IT Services agency is working with federal and state partners to investigate and fortify the security of the state's critical infrastr…

Jul 29·bleepingcomputer.com

Your AI Agents Are Guessing at Scale: Permissions Decide the Damage

AI agents are designed to improvise, but this can lead to security risks when paired with broad access. Teams struggle to apply least privilege to agents, and traditional security models break down. Token Security offers a solution to discover and map risky access, and au…