Case shai-hulud-tensorlake-npmAI supply chainWorldwide
Shai-Hulud worm compromises Tensorlake npm release
A Shai-Hulud (ChainDrop-linked) worm variant reached AI agent infrastructure when tensorlake npm SDK version 0.5.144 was published through a compromised maintainer account. Socket flagged the credential-stealing payload 11 minutes after publication, the version was live for about 1 hour 40 minutes before removal, and the impact is unknown.
MediumAttack on AIStatus: contained; impact unknownVerified case file
Open in the interactive mapCite this case
- Reported
- Oct 8, 2026
- Location
- Worldwide
- Target
- Tensorlake AI agent platform npm SDK (tensorlake 0.5.144) and developers who installed it
- Severity
- Medium
- Verification
- Reported by credible press or researchers
- Triggered by
- Unknown (Shai-Hulud / ChainDrop-linked worm variant)
- Actor type
- Unknown
- Sector
- Technology & AI vendors
- Reported loss
- None reported
- CVE
- None
What happened
A Shai-Hulud (ChainDrop-linked) worm variant reached AI agent infrastructure when tensorlake npm SDK version 0.5.144 was published through a compromised maintainer account. Socket flagged the credential-stealing payload 11 minutes after publication, the version was live for about 1 hour 40 minutes before removal, and the impact is unknown.
Impact: Impact unknown; the malware was detected within minutes of release.
Who triggered it
Actor: Unknown (Shai-Hulud / ChainDrop-linked worm variant)
Attribution: Linked to the Shai-Hulud worm family; the operators are not named.
Motive: Credential theft
How it happened
Attack vector: Malicious npm release of a Tensorlake package carrying credential-stealing malware.
Root cause: Unknown; the Tensorlake compromise let the worm publish a malicious npm release.
Timeline
- Reported by The Register
Response
Detection within minutes of the npm release; no other response details given.
Current status: contained; impact unknown
Research warnings this case matches
- Seen in real incidents AI systems themselves become targets · International AI Safety Report
- Seen in real incidents Privacy and security · MIT AI Risk Repository
Compared with published AI risk research from several institutes on the Research vs reality page.
Sources (open in a new tab)
- The Register theregister.com
- Socket socket.dev
- Tensorlake GitHub security advisory github.com
Cite this case
Shai-Hulud worm compromises Tensorlake npm release. AI Threat Watch, case shai-hulud-tensorlake-npm, Oct 8, 2026. https://aithreat.io/cases/shai-hulud-tensorlake-npm/
Permanent link: https://aithreat.io/cases/shai-hulud-tensorlake-npm/
Related cases
The same type of incident, most recent first.