Shai-Hulud: npm and PyPI Cross-Ecosystem Worm

The Shai-Hulud (ChainDrop) campaign, attributed to TeamPCP, employs a self-propagating worm targeting the npm and PyPI ecosystems. By compromising npm packages such as openapi-react-query, the malware exfiltrates environment variables and registry API tokens from developer workstations. It uniquely implements a cross-ecosystem pivot, utilizing stolen PyPI tokens to autonomously publish malicious Python packages, creating a recursive infection loop. This campaign affected over 400 npm packages with an aggregate reach of 59 million monthly downloads, demonstrating a scalable, automated supply chain attack vector that bypasses traditional static trojan limitations through autonomous credential harvesting and republishing.

• Attack Vector: Trojanized npm Distribution

  • Initial access achieved via malicious releases of libraries including openapi-react-query and 7nohe-openapi-react-query-codegen.
  • Malware executes upon installation to scan local environments for .env files, cloud provider secrets (AWS/GCP/Azure), and registry authentication tokens.
  • Specifically targets developer workstations to secure high-privilege tokens required for unauthorized registry uploads.

• Technical Mechanics: Autonomous Cross-Ecosystem Pivot

  • Employs a self-replicating engine that automates package version bumping and republishing without manual C2 intervention.
  • Bridges the gap between language environments by identifying PyPI tokens on machines where developers maintain both JavaScript and Python projects.
  • Creates a compounding infection cycle where compromised maintainer accounts serve as autonomous launchpads for further supply chain poisoning.

• Threat Actor Profile: TeamPCP Orchestration

  • Attributed to TeamPCP, a group characterized by high-scale automation and sophisticated supply chain orchestration.
  • Evidence suggests the use of autonomous agents to handle target selection and execution logic during the replication phase.
  • Utilizes a structured, automated offensive methodology designed for maximum dissemination speed and reach.

• Scale and Ecosystem Impact

  • Identified over 400 compromised npm packages acting as nodes in the propagation chain.
  • Entry-point packages reached up to 460,000 weekly downloads, facilitating massive downstream exposure.
  • Aggregate impact estimated at 59 million monthly downloads, exposing a critical risk in shared developer credential management across registries.

• Defensive Actions and Mitigation

  • Immediate rotation of all npm, PyPI, and cloud provider API tokens and secrets across affected development teams.
  • Implementation of strict dependency pinning and checksum verification to prevent the installation of unauthorized version bumps.
  • Deployment of runtime monitoring and secret scanning to detect and block unauthorized registry uploads originating from developer workstations.