TanStack npm supply chain attack now a full campaign, targets AI developer tooling

International Cyber Digest X/Twitter thread reporting an escalation of the TanStack npm supply chain attack into a broader campaign. Dated May 12, 2026. This is a social media post: treat as tip/investigation lead pending corroboration.

Source Summary

What It Contains

International Cyber Digest X/Twitter thread reporting an escalation of the TanStack npm supply chain attack into a broader campaign. Dated May 12, 2026. This is a social media post: treat as tip/investigation lead pending corroboration.

Extracted Claims

  • Campaign name escalation: The original TanStack npm supply chain attack (May 12 2026, 42 official tanstack packages) has escalated to a full campaign under the moniker "'Mini' Shai-Hulud"
  • Newly targeted organizations/platforms: OpenSearch, Mistral AI, Guardrails AI, UiPath, and "Squawk packages" across both npm and PyPI
  • Malware name: "Shai-Hulud" (parent campaign), "Mini" Shai-Hulud (scaled variant)
  • Target: AI developer tooling ecosystem
  • Persistence mechanism: The malware hooks into Claude Code configuration (.claude/settings.json) and VS Code tasks (.vscode/tasks.json) to re-execute on every tool event, long after the infected package is removed. npm uninstall does not remediate.
  • Dead-man's switch (original attack): Payload plants a watcher that destroys the user's home directory the moment the stolen GitHub token is revoked
  • Scale: 42 official tanstack npm packages were initially compromised (May 12 2026)

Key Claims (Unverified: Secondary Source)

  • Cross-platform attack (npm + PyPI) targeting OpenSearch, Mistral AI, Guardrails AI, UiPath, Squawk
  • Persistence via Claude Code and VS Code config file manipulation
  • npm uninstall insufficient for remediation
  • Dead-man's switch destroys home directory on GitHub token revocation

Evidence Quality

Social media post from a cybersecurity newsletter account. Claims are specific and internally consistent. Corroboration from additional sources recommended before treating as confirmed. Trust level: secondary: useful as a lead and for timeline tracking.

Follow-Up

  • Search for additional sources: any blog posts, vendor analyses, or CVE disclosures related to TanStack npm attack, Shai-Hulud, or cross-platform AI tooling supply chain compromises
  • Cross-reference with npm/PyPI package maintainer advisories
  • Corroborate with primary advisories or vendor writeups, then update the linked canonical finding with confirmed package names, versions, IOCs, and confidence