CVE-2026-6859HIGH 8.8EPSS p27.1%
CVE-2026-6859CVE-2026-6859
Description
A flaw was found in InstructLab. The `linux_train.py` script hardcodes `trust_remote_code=True` when loading models from HuggingFace. This allows a remote attacker to achieve arbitrary Python code execution by convincing a user to run `ilab train/download/generate` with a specially crafted malicious model from the HuggingFace Hub. This vulnerability can lead to complete system compromise.
Scoring
| CVSS 3.1 | 8.8 (HIGH) |
| Vector | CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H |
| EPSS | 0.35% probability of exploitation · percentile 27.1% · 2026-06-19T12:03:05Z |
| Published | 2026-04-22 |
| Last modified | 2026-05-06 |
Underlying weaknesses· 1
References
1
| Type | Target | Confidence | Tier |
|---|---|---|---|
| Weakness | Inclusion of Functionality from Untrusted Control Spherecwe-829 | 0% | live |
Related by meaning· 6
Nearest entities by semantic similarity across the cs-graph corpus.