RSS
Last updated: May 13, 2026 at 5:42 AM UTC
All 208 Vulnerability 72 Breach 41 Threat 88 Defense 7
Tag: ai-ml (1 article)Clear

The same supply-chain worm that hit SAP packages on Wednesday spread to PyTorch Lightning and Intercom's npm SDK on Thursday

Update on the Mini Shai-Hulud campaign covered April 30: The same supply-chain worm that hit four SAP npm packages on Wednesday spread to two more major packages on Thursday. PyTorch Lightning, an AI training framework with 31,100 GitHub stars and hundreds of thousands of daily downloads, had malicious versions 2.6.2 and 2.6.3 published on PyPI for 42 minutes before being quarantined. Intercom-client, the official Node.js SDK for Intercom (361,510 weekly downloads), was compromised at 14:41 UTC. Intercom traced its compromise to pyannote-audio pulling Lightning as a dependency - showing the worm propagating through stolen credentials from the SAP victims.

Check
Audit any developer machine or CI runner that ran 'pip install' on PyTorch Lightning or 'npm install' on intercom-client between April 30 and May 1, and rotate every credential on those machines.
Affected
Lightning (PyPI) versions 2.6.2 and 2.6.3 - safe version is 2.6.1. Intercom-client (npm) version 7.0.4 (per Socket) and 7.0.5 (per Wiz). AI/ML environments running Lightning routinely hold GPU cluster credentials, cloud IAM tokens, Hugging Face API keys, and Weights & Biases tokens. Backend services and CI/CD pipelines integrating with Intercom's API are exposed even if they don't use Lightning.
Fix
Pin Lightning to 2.6.1 or earlier; reject 2.6.2 and 2.6.3. Update intercom-client per Intercom's advisory. Rotate all credentials potentially exposed: GitHub tokens, npm tokens, AWS/GCP/Azure keys, environment-variable secrets. Gate npm publish behind environment review (the same pattern that compromised SAP).