MAL-2026-17457
Vulnerability from ossf_malicious_packages
-= Per source details. Do not edit below this line.=-
Source: amazon-inspector (4b542a4c53b4c08e104eebfade64c8e91c8d731c4a9bd298e48a258117c5b1bb)
The package advertises itself as a TTS evaluation harness but ships two large bundled binaries and executes them on the installer's host. voxeval/bin/voxtts-core is an ~8MB ELF launched with an XMRig-shaped configuration (pools, donate-level, print-time, huge-pages, cpu.threads, user=run_id, pass="vox") pointing at a hardcoded Monero pool at 185.194.177.249:443. The pool host is split into two base64 chunks (MTg1LjE5NA==, MTc3LjI0OQ==) and reassembled at runtime, and configuration keys are string-concatenated ("po"+"ools", "--"+"config=") to defeat literal keyword scans. A second bundled binary, a ~40MB cloudflared, is invoked as cloudflared access tcp --hostname <edge> --url 127.0.0.1:<port> --no-autoupdate to proxy the miner's egress through Cloudflare's edge and evade DNS/IP-based blocking of the pool. A _Progress helper fabricates fake training-progress lines (step N/250000 loss=... tok/s=...) so the miner's output resembles ML training in notebook logs. The dropper chmods the core binary to 0o755 and spawns it via Popen. Installing or invoking voxeval causes sustained CPU consumption on the installer's host for the benefit of the attacker's mining wallet, with egress deliberately obscured.
Source: kam193 (4ffc5162b3b77de40988aa702c4493260a4f0b4f09d5f4719feabcd69f33908a)
The package imitates real activity and instead deploys a coin miner.
Category: MALICIOUS - The campaign has clearly malicious intent, like infostealers.
Campaign: 2026-10-voxeval
Reasons (based on the campaign):
- cryptominer
- CWE-506 - The product contains code that appears to be malicious in nature.
- CWE-506 - The product contains code that appears to be malicious in nature.
- CWE-506 - The product contains code that appears to be malicious in nature.
- CWE-506 - The product contains code that appears to be malicious in nature.
{
"affected": [
{
"database_specific": {
"cwes": [
{
"cweId": "CWE-506",
"description": "The product contains code that appears to be malicious in nature.",
"name": "Embedded Malicious Code"
},
{
"cweId": "CWE-506",
"description": "The product contains code that appears to be malicious in nature.",
"name": "Embedded Malicious Code"
},
{
"cweId": "CWE-506",
"description": "The product contains code that appears to be malicious in nature.",
"name": "Embedded Malicious Code"
},
{
"cweId": "CWE-506",
"description": "The product contains code that appears to be malicious in nature.",
"name": "Embedded Malicious Code"
}
],
"indicators": {
"evidence_files": [
{
"path": "voxeval/harness.py",
"sha256": "ec0af743d57a732705ff82edd8f198980e021e99b2351fc24c7dc05fd0d4b0e2",
"tlsh": "9fd182f5d9161a62c3d7da124946b593630bac43974e08387cfc83182fa4674c6f1efa"
}
],
"package_integrity": [
{
"filename": "voxeval-0.4.5-py3-none-any.whl",
"hashes": {
"blake2b_256": "569535aff70206d64220197c2e6101ece0d65de160fd8acf86e3969fd0c5b744",
"md5": "212526c3ee52e82f273794455e409cda",
"sha256": "ecb5b0a037f57477c3173c4e6a28381b62a46ec8193430c54c36258fa1a920a8"
}
}
]
}
},
"package": {
"ecosystem": "PyPI",
"name": "voxeval"
},
"versions": [
"0.4.2",
"0.4.3",
"0.4.4",
"0.4.5"
]
}
],
"credits": [
{
"contact": [
"inspector-research@amazon.com"
],
"name": "Amazon Inspector",
"type": "FINDER"
},
{
"contact": [
"https://github.com/kam193",
"https://bad-packages.kam193.eu/"
],
"name": "Kamil Ma\u0144kowski (kam193)",
"type": "REPORTER"
}
],
"database_specific": {
"malicious-packages-origins": [
{
"id": "pypi/2026-10-voxeval/voxeval",
"import_time": "2026-10-02T21:41:25.235176359Z",
"modified_time": "2026-10-02T21:28:42.41347Z",
"sha256": "4ffc5162b3b77de40988aa702c4493260a4f0b4f09d5f4719feabcd69f33908a",
"source": "kam193",
"versions": [
"0.4.2",
"0.4.3",
"0.4.4",
"0.4.5"
]
},
{
"id": "pypi/2026-10-voxeval/voxeval",
"import_time": "2026-10-02T22:41:28.867292387Z",
"modified_time": "2026-10-02T22:25:09.940191Z",
"sha256": "b9e9f292852286c6537dcd1e5d4047dc3126a413f7d65385b54304156f2b6a83",
"source": "kam193",
"versions": [
"0.4.2",
"0.4.3",
"0.4.4",
"0.4.5"
]
},
{
"id": "pypi/2026-10-voxeval/voxeval",
"import_time": "2026-10-02T23:39:22.663770673Z",
"modified_time": "2026-10-02T23:19:56.752864Z",
"sha256": "a489350e0f4586ebe4658874a085f3f100ada8bb1018a2b603b624ee0516c58f",
"source": "kam193",
"versions": [
"0.4.2",
"0.4.3",
"0.4.4",
"0.4.5"
]
},
{
"id": "IN-MAL-2026-020932",
"import_time": "2026-10-04T23:40:43.359281043Z",
"modified_time": "2026-10-04T23:21:53Z",
"sha256": "4b542a4c53b4c08e104eebfade64c8e91c8d731c4a9bd298e48a258117c5b1bb",
"source": "amazon-inspector",
"versions": [
"0.4.5"
]
},
{
"id": "IN-MAL-2026-020930",
"import_time": "2026-10-04T23:40:43.134954246Z",
"modified_time": "2026-10-04T23:21:32Z",
"sha256": "70fb2627789c646979bbe14294d6e5c8922698cbbec8937944a2c1fe62304874",
"source": "amazon-inspector",
"versions": [
"0.4.4"
]
},
{
"id": "IN-MAL-2026-020929",
"import_time": "2026-10-04T23:40:43.045706428Z",
"modified_time": "2026-10-04T23:21:21Z",
"sha256": "7ef0f5b96d43ba8e51ec9c8fd4d48e3df446414b6c38170f2a002c724ca603ad",
"source": "amazon-inspector",
"versions": [
"0.4.3"
]
},
{
"id": "IN-MAL-2026-020928",
"import_time": "2026-10-04T23:40:42.948791335Z",
"modified_time": "2026-10-04T23:21:10Z",
"sha256": "bb1ba93c8d397330d4cb7ce50d874c48634ae76f7aec84a8d62fd04f8958509d",
"source": "amazon-inspector",
"versions": [
"0.4.2"
]
}
]
},
"details": "\n---\n_-= Per source details. Do not edit below this line.=-_\n\n## Source: amazon-inspector (4b542a4c53b4c08e104eebfade64c8e91c8d731c4a9bd298e48a258117c5b1bb)\nThe package advertises itself as a TTS evaluation harness but ships two large bundled binaries and executes them on the installer\u0027s host. voxeval/bin/voxtts-core is an ~8MB ELF launched with an XMRig-shaped configuration (pools, donate-level, print-time, huge-pages, cpu.threads, user=run_id, pass=\"vox\") pointing at a hardcoded Monero pool at 185.194.177.249:443. The pool host is split into two base64 chunks (`MTg1LjE5NA==`, `MTc3LjI0OQ==`) and reassembled at runtime, and configuration keys are string-concatenated (`\"po\"+\"ools\"`, `\"--\"+\"config=\"`) to defeat literal keyword scans. A second bundled binary, a ~40MB `cloudflared`, is invoked as `cloudflared access tcp --hostname \u003cedge\u003e --url 127.0.0.1:\u003cport\u003e --no-autoupdate` to proxy the miner\u0027s egress through Cloudflare\u0027s edge and evade DNS/IP-based blocking of the pool. A `_Progress` helper fabricates fake training-progress lines (`step N/250000 loss=... tok/s=...`) so the miner\u0027s output resembles ML training in notebook logs. The dropper chmods the core binary to 0o755 and spawns it via Popen. Installing or invoking voxeval causes sustained CPU consumption on the installer\u0027s host for the benefit of the attacker\u0027s mining wallet, with egress deliberately obscured.\n\n## Source: kam193 (4ffc5162b3b77de40988aa702c4493260a4f0b4f09d5f4719feabcd69f33908a)\nThe package imitates real activity and instead deploys a coin miner.\n\n\n---\n\nCategory: MALICIOUS - The campaign has clearly malicious intent, like infostealers.\n\n\nCampaign: 2026-10-voxeval\n\n\nReasons (based on the campaign):\n\n\n - cryptominer\n",
"id": "MAL-2026-17457",
"modified": "2026-10-04T23:42:41.657101664Z",
"published": "2026-10-02T21:28:42Z",
"references": [
{
"type": "WEB",
"url": "https://bad-packages.kam193.eu/pypi/package/voxeval"
},
{
"type": "PACKAGE",
"url": "https://pypi.org/project/voxeval/0.4.5/"
},
{
"type": "PACKAGE",
"url": "https://pypi.org/project/voxeval/0.4.4/"
},
{
"type": "PACKAGE",
"url": "https://pypi.org/project/voxeval/0.4.3/"
},
{
"type": "PACKAGE",
"url": "https://pypi.org/project/voxeval/0.4.2/"
}
],
"schema_version": "1.7.4",
"summary": "Malicious code in voxeval (PyPI)"
}
Drag nodes to arrange them. Double-click a related vulnerability to open it, or a source to visit its link. The Physics button on the left tunes the layout and the simulation; the legend at the bottom left hides or shows nodes by kind and by sighting type.
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
Vulnerability timeline
Sightings and recorded dates, grouped by day · UTC
Loading timeline…
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.