Remote code execution via unsafe model-output eval
Published Sep 4, 2026 · Updated Sep 8, 2026
Code injection in LaVague 0.2.35 and earlier allows remote attackers to execute arbitrary Python code through malicious web-page content. PythonFromMarkdownExtractor.extract_as_object extracts a Python block from model output and passes it directly to eval without restricting expressions or built-ins. Exploitation requires an agent workflow to process attacker-influenced page content, after which the code runs on the operator host without per-action review.
Summary
What happened
Code injection in LaVague 0.2.35 and earlier allows remote attackers to execute arbitrary Python code through malicious web-page content. PythonFromMarkdownExtractor.extract_as_object extracts a Python block from model output and passes it directly to eval without restricting expressions or built-ins. Exploitation requires an agent workflow to process attacker-influenced page content, after which the code runs on the operator host without per-action review.
The record
- CVE
- CVE-2026-85694
- Published
- Sep 4, 2026
- Updated
- Sep 8, 2026
- Vendor
- LaVague AI
- Product
- LaVague
- Classifications
- CWE-94, T1059.006
- Attack vector
- network
- Privileges
- unauthenticated
Timeline
How it unfolded
- Sep 4, 2026CVE publishedPublication date reported by the CVE source.
- Sep 8, 2026Record updatedLatest update available in the CVE record.
Exploitability
Present is not the same as exploitable
Compare your product and version with the public record. A matching version still requires validation against your environment.
Is a vulnerable build present?
Compare these published version ranges with your installed build and any vendor patches.
- Affected versionversion=0 <=0.2.35
What conditions does exploitation require?
What is affected?
Attacks
What attackers are doing with it
Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.
Weakness, pattern, technique
Public exploit references
- LaVague issue 650 command-execution demonstrationproof of concept · demonstrated
Labels summarize the accepted research assessment. They do not indicate a test against your environment.
Technologies
Your stack
See the directory against your own environment.
Your stack
Check the software in your environment
Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.
Book a demo