Skip to content

CVE-2024-6331 Injection by Prompt Injection in stitionai/devika

MediumCVSS 6.5 EPSS 0.50%
Published
Updated
Exploit status
No public exploit known
CISA SSVC
Proof of concept public · automatable: no · technical impact: partial

Description

stitionai/devika main branch as of commit cdfb782b0e634b773b10963c8034dc9207ba1f9f is vulnerable to Local File Read (LFI) by Prompt Injection. The integration of Google Gimini 1.0 Pro with HarmBlockThreshold.BLOCK_NONE for HarmCategory.HARM_CATEGORY_HATE_SPEECH and HarmCategory.HARM_CATEGORY_HARASSMENT in safety_settings disables content protection. This allows malicious commands to be executed, such as reading sensitive file contents like /etc/passwd.

Upgrade path

2 packages: 1 fixed, 1 partially fixed.

PackageAffectedFixed inStatus
stitionai devika <= latest; >= 2024-05-02 after latest Partially fixed
stitionai/devika <= latest after latest Fixed

Affected products

ProductAffected versionsFixed in
stitionai/devika <= latest No fix listed
stitionai devika <= latest No fix listed
stitionai devika >= 2024-05-02 No fix listed

References