Which versions of kaggle-runner are malicious?

The following PyPI versions of kaggle-runner are flagged malicious: 0.0.2. Treat any of these as compromised.

How do I remediate kaggle-runner if I installed it?

kaggle-runner is built to steal secrets, so assume every credential the build or runtime could read is compromised. Remove it from your project and lockfile, then rotate ALL exposed secrets — npm/registry tokens, cloud keys, CI/CD secrets, SSH keys, and any .env values — from a known-clean machine. Audit logs for unauthorized use of those credentials.

I already installed kaggle-runner — how do I know if it already compromised me?

If kaggle-runner was ever installed, its post-install/runtime payload may have already executed. O3's L7 egress monitoring and runtime eBPF sensors detect the credential exfiltration or command-and-control callback after install and block the malicious outbound channel, so you catch and contain the actual compromise — not just the presence of the package.

Is kaggle-runner part of a known attack campaign?

Yes — kaggle-runner is associated with the IN-MAL-2026-002524 campaign. Packages in the same campaign often share infrastructure and payloads, so check whether any related packages are also in your dependency tree.

How do I find kaggle-runner across my projects, lockfiles, and CI pipelines?

Scan your lockfiles (package-lock.json, pnpm-lock.yaml, yarn.lock, requirements.txt, poetry.lock, etc.) and build artifacts for kaggle-runner (version 0.0.2). O3 Security's supply-chain scanner checks every dependency against known-malicious package intelligence at install time and in CI, flagging kaggle-runner across your stack and pipelines.

Malicious package

kaggle-runnerPyPI

Malicious code in kaggle-runner (PyPI) Remove it immediately and rotate any exposed credentials.

MAL-2026-3693

Immediate action

Remove the package, then rotate any secrets the build/runtime could reach.

pip uninstall kaggle-runner

What this malware does

kaggle_runner/coordinator.py embeds a bash reverse-shell template (rvs_str) that connects to vtool.duckdns.org:23454 via ncat with retry/backoff plus a heartbeat channel on port 23455. When a consumer calls Coordinator.create_runner(config), the package writes rvs.sh alongside entry.sh/runner.sh/setup_pty/gdrive_setup into a kernel folder; Coordinator.run_local() then executes python main.py, which invokes bash -x entry.sh, which in turn backgrounds rvs.sh — opening an interactive shell from the runner's host back to the author-controlled duckdns.org subdomain. The same bundle wgets a gdrive binary from github.com/gdrive-org/gdrive/releases/download/2.1.0/gdrive-linux-x64 and installs it to /bin/gdrive. None of this behavior is documented in the README (which advertises AMQP logging for Kaggle kernels). The reverse shell does not fire at import/install time — setup.py and init.py are clean — but it fires as part of the package's advertised Coordinator API flow, so any consumer who actually uses the library exposes the executing host (their machine or a Kaggle kernel they push) to the author. A separate file (kaggle_runner/utils/utils.py) also hardcodes CloudAMQP credentials (termite.rmq.cloudamqp.com / drdsfaew) with a comment 'oh~ just give my password out~' — this is author self-harm and on its own would be allow, but combined with the reverse-shell pipe to a duckdns C2 host, the installer-side impact is clear.

Malicious versions

1 flagged

0.0.2

Indicators of compromise (SHA-256)

8dcd49ca70b987b236ba4341d839addfec9afb344e1471195f2f825281092f71

Detection & response playbook

Credential / info stealer

Find it
Scan your lockfiles (package-lock.json, pnpm-lock.yaml, yarn.lock, requirements.txt, poetry.lock, etc.) and build artifacts for kaggle-runner (version 0.0.2). O3 Security's supply-chain scanner checks every dependency against known-malicious package intelligence at install time and in CI, flagging kaggle-runner across your stack and pipelines.
If you installed it — respond
kaggle-runner is built to steal secrets, so assume every credential the build or runtime could read is compromised. Remove it from your project and lockfile, then rotate ALL exposed secrets — npm/registry tokens, cloud keys, CI/CD secrets, SSH keys, and any .env values — from a known-clean machine. Audit logs for unauthorized use of those credentials.
Did it already run?
If kaggle-runner was ever installed, its post-install/runtime payload may have already executed. O3's L7 egress monitoring and runtime eBPF sensors detect the credential exfiltration or command-and-control callback after install and block the malicious outbound channel, so you catch and contain the actual compromise — not just the presence of the package.
How O3 protects you
O3 blocks kaggle-runner before install through its supply-chain scanner, and if it has already run, detects and severs the exfiltration or C2 callback at runtime through L7 egress monitoring and eBPF.

Frequently asked questions

No. kaggle-runner on PyPI has been identified as a malicious package (version 0.0.2 flagged). It should be removed immediately — do not install or keep it in your dependency tree.

Campaign

IN-MAL-2026-002524

References

pypi.org

Credits

Amazon Inspector · finder

Detect & block this

O3 blocks kaggle-runner-class packages before install and in CI — and if it already ran, its runtime egress monitoring catches the credential exfiltration and severs the channel.

Malware detection Runtime egress monitoring