Supply chain · pip · High

Critical Malicious 'cognikit' Package in PyPI Steals Credentials and Enables Remote Control

Threat Engine Auto-Feed · data current as of 2026-07-31

A critical malicious package, 'cognikit' version 0.1.2, has been identified in the PyPI ecosystem. This package is designed for credential and secret theft, exfiltrating browser data, modifying cryptocurrency wallet addresses, and establishing remote control, posing a significant supply-chain risk to users who installed it.

  • Package: pip / cognikit
  • Affected versions: = 0.1.2
  • Severity: critical
  • Reach: unknown
  • Exposure: 65/100 (High)

What happened — Credential / secret theft

The 'cognikit' package (version 0.1.2) in PyPI contains malicious code. It is designed to be pulled as a dependency, providing C2 configuration and malicious functionality. This package is associated with a campaign that includes other packages providing fake functionality while silently executing malicious actions from dependencies.

How the attack works

This supply-chain attack involves a malicious package being introduced into a public repository. When installed, the package executes code to exfiltrate sensitive data, modify clipboard contents (specifically cryptocurrency wallet addresses), and establish persistent remote access to the compromised system. The attack leverages dependencies to hide its true intent, making it difficult to detect during initial package review.

Who's exposed

Profiles most at risk

  • Developers and organizations using the PyPI ecosystem
  • Teams whose build processes automatically pull the latest versions of dependencies
  • Users involved in cryptocurrency projects or those undergoing technical assessments that might use these packages as dependencies

Conditions that increase exposure

  • Unpinned dependencies in project configurations
  • Lack of dependency lockfiles (e.g., requirements.txt with exact versions)
  • Use of internal package mirrors that do not vet upstream packages
  • Automated CI/CD pipelines that install dependencies without strict versioning

Blast-radius scenarios

  • An attacker gains remote control over the compromised device, enabling further malicious activities.
  • Sensitive browser data, including credentials and personal information, is exfiltrated.
  • Cryptocurrency transactions are redirected due to modified wallet addresses in the clipboard.
  • Persistence mechanisms are established, allowing continued access to the compromised system.

What to do (defensive)

Detect

  • Scan project dependencies for 'cognikit' version 0.1.2.
  • Review network traffic for suspicious C2 communications from development or production environments.
  • Monitor for unexpected file system changes or new processes initiated by Python environments.
  • Utilize software composition analysis (SCA) tools to identify known malicious packages.

Contain

  • Immediately isolate any systems found to have installed 'cognikit' version 0.1.2.
  • Block network access to known C2 infrastructure if identified.
  • Revoke and rotate any credentials or API keys that may have been exposed on compromised systems.
  • Implement strict egress filtering to prevent unauthorized data exfiltration.

Remediate

  • Remove 'cognikit' version 0.1.2 from all affected projects and environments.
  • Perform a clean reinstallation of all dependencies after verifying their integrity.
  • Force password resets for all users and services that may have been exposed.
  • Restore affected systems from known clean backups, if available.
  • Implement dependency pinning and lockfiles to prevent automatic installation of new malicious versions.

Frontier verdict — High

Prioritize immediate detection and removal of 'cognikit' version 0.1.2 due to its critical severity and capabilities for credential theft, remote control, and cryptocurrency redirection.

Sources

For detection-engineering and awareness only · point-in-time · not security advice · sourced from the GitHub Advisory Database.