@whalent/agent-corenpm
Advisory published Updated
@whalent/agent-core is a confirmed malicious npm package (MAL-2026-10722) that executes malicious code on install (malicious versions 0.3.230, 0.3.231, 0.3.232…). Do not install it — remove it immediately and rotate any exposed credentials.
Malicious code in @whalent/agent-core (npm)
What this malware does
The package's main/bin entry (dist/index.cjs) implements a remote agent that opens a WebSocket to a gateway configured via WHALENT_GATEWAY / WHALENT_TOKEN / WHALENT_PLATFORM_URL and, over that channel, drives local PTY sessions via node-pty, spawns shells against process.env.SHELL, bridges loopback services on 127.0.0.1:3389 (RDP) and 127.0.0.1:5900/5901 (VNC), and runs a Python Jupyter kernel sidecar (dist/jupyter_sidecar.py) that executes arbitrary NDJSON-delivered code cells. The remote party at the gateway therefore has interactive command and code execution on the installer's host, plus a channel to reach otherwise-loopback-bound RDP/VNC. The 13.6 MB dist/index.cjs is protected with obfuscator.io-style transforms — a rotated 105,426-entry string array, _0xNNNN identifier renaming, decoder wrappers, and control-flow flattening — hiding the gateway URLs, command dispatch, and upgrade logic from review. The bundle additionally contains an npmInstall code path plus WHALENT_NPM_REGISTRY and WHALENT_CORE_ENTRY environment configuration, and the README describes the package as one that is 'replaced by remote daemon upgrades', so the running agent can be instructed by the gateway to fetch and swap in a new version of @whalent/agent-core from a configurable registry, executing whatever code that fetched artifact contains. The net effect of installing and running this package is a persistent, remote-controlled RCE channel into the installer's machine with a self-update mechanism that can pull further code from a registry the remote party can influence.
Malicious versions
Indicators of compromise (SHA-256)
Detection & response playbook
Malicious packageFind it
Search your lockfiles and build artifacts for @whalent/agent-core (49 malicious versions).
If you installed it — respond
Remove @whalent/agent-core 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 @whalent/agent-core 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 @whalent/agent-core-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.