---
id: CVE-2026-122160
title: >-
  myshell-ai AIlice HuggingFace Model Wrapper AModelCausalLM.py LoadModel_PEFT
  inclusion of functionality from untrusted control sphere
summary: >-
  A security vulnerability has been detected in myshell-ai AIlice up to 0.4.0.
  This vulnerability affects the function LoadModel_PEFT of the file
  ailice/core/llm/AModelCausalLM.py of the component HuggingFace Model Wrapper.
  The manipulatio…
severity: medium
cvss: 6.3
cvssVector: 'CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:L/I:L/A:L/E:P/RL:X/RC:C'
cvssSource: cna
cwe:
  - CWE-829
vendor: myshell-ai
product: AIlice
affected:
  - AIlice 0.1
  - AIlice 0.2
  - AIlice 0.3
  - AIlice 0.4.0
ssvc:
  exploitation: poc
  automatable: 'no'
  technicalImpact: partial
  timestamp: '2026-11-05T20:31:23.004757Z'
exploitAvailable: true
published: '2026-11-05'
updated: '2026-11-05'
sourceUpdated: '2026-11-05T20:31:37.053Z'
source: CVEORG
sourceUrl: 'https://www.cve.org/CVERecord?id=CVE-2026-122160'
references:
  - url: 'https://vuldb.com/vuln/431160'
    label: >-
      VDB-431160 | myshell-ai AIlice HuggingFace Model Wrapper AModelCausalLM.py
      LoadModel_PEFT inclusion of functionality from untrusted control sphere
  - url: 'https://vuldb.com/vuln/431160/cti'
    label: 'VDB-431160 | CTI Indicators (IOB, IOC, IOA)'
  - url: 'https://vuldb.com/cve/CVE-2026-122160'
    label: CVE-2026-122160 | CVE Analysis and Report
  - url: 'https://vuldb.com/submit/1028200'
    label: >-
      Submit #1028200 | MyShell AIlice 0.4.0 (commit
      63a22105cdf8c0d297af332010f46e4153bbc1ec) Improper Input Validation
  - url: 'https://github.com/myshell-ai/AIlice/issues/77'
  - url: 'https://github.com/user-attachments/files/33392335/poc.zip'
  - url: 'https://github.com/myshell-ai/AIlice/'
tags:
  - cve.org
  - exploit-available
epss: 0.00312
epssPercentile: 0.24188
ingestedAt: '2026-11-05T22:38:31.838Z'
---

## Overview

A security vulnerability has been detected in myshell-ai AIlice up to 0.4.0. This vulnerability affects the function LoadModel_PEFT of the file ailice/core/llm/AModelCausalLM.py of the component HuggingFace Model Wrapper. The manipulation of the argument base_model_name_or_path leads to inclusion of functionality from untrusted control sphere. It is possible to initiate the attack remotely. The exploit has been disclosed publicly and may be used. The project was informed of the problem early through an issue report but has not responded yet.

## Affected

- `AIlice 0.1`
- `AIlice 0.2`
- `AIlice 0.3`
- `AIlice 0.4.0`

## Remediation

Refer to the linked advisories for vendor-supplied fixes and affected version ranges.