tensorlakenpm
Advisory published Updated
tensorlake is a confirmed malicious npm package (MAL-2026-17650) that executes malicious code on install (malicious version 0.5.144). Do not install it — remove it immediately and rotate any exposed credentials.
Malicious code in tensorlake (npm)
What this malware does
The npm package [email protected] ships lib/setup.mjs as a heavily obfuscated single-line script using hex-mangled identifiers (_0x15c964, _0x35672e,...), a transformation shape typical of hostile loader code rather than ordinary minification. Multiple shipped bundles under dist/ (dist/index.cjs, dist/applications/index.cjs, dist/sandbox-image.cjs, dist/function-agent/main.cjs, dist/function-executor/main.cjs, and their.js counterparts) combine require("child_process") with repeated POST request primitives and shell utilities such as curl and ping, including host/id-collection patterns alongside outbound HTTP calls. Automated contextual tracing of these files was blocked by the provider safety filter after the model engaged with the content, which is a pattern that correlates with hostile code rather than benign SDK traffic. The combination of an obfuscated setup script and child_process + hardcoded HTTP POST patterns in the shipped bundles is consistent with install-time or load-time execution of attacker-controlled behavior against the installer's host.
Any computer that has this package installed or running should be considered fully compromised. All secrets and keys stored on that computer should be rotated immediately from a different computer. The package should be removed, but as full control of the computer may have been given to an outside entity, there is no guarantee that removing the package will remove all malicious software resulting from installing it.
Malicious versions
Indicators of compromise (SHA-256)
Detection & response playbook
Malicious packageFind it
Search your lockfiles and build artifacts for tensorlake (version 0.5.144).
If you installed it — respond
Remove tensorlake from your project and lockfile, then assume any secrets accessible to the build or runtime were exposed: rotate API keys, tokens, and credentials, and audit for unexpected outbound activity or persistence.
Did it already run?
If tensorlake was installed, its post-install payload may already have run. Removing the package does not undo that — check outbound connections and credential use from the install window onward.
Frequently asked questions
Campaign
References
Credits
- Amazon Inspector · finder
Detect & block this
O3 blocks tensorlake-class packages before install and in CI — and if it already ran, its runtime egress monitoring catches the malicious outbound activity and severs the channel.