Authenticated command execution via unsandboxed generated Python
Published Mar 20, 2026 · Updated Mar 20, 2026
Code injection in Bag of Words 0.0.297 allows remote authenticated users to execute arbitrary commands through an AI-generated function. The generate_df path executes Python produced by an LLM without AST validation, so crafted prompt text can induce operating-system command calls on the server. An account able to query a configured LLM and data source is required, and successful exploitation runs commands with the application worker's privileges.
Summary
What happened
Code injection in Bag of Words 0.0.297 allows remote authenticated users to execute arbitrary commands through an AI-generated function. The generate_df path executes Python produced by an LLM without AST validation, so crafted prompt text can induce operating-system command calls on the server. An account able to query a configured LLM and data source is required, and successful exploitation runs commands with the application worker's privileges.
The record
- CVE
- CVE-2026-4500
- Published
- Mar 20, 2026
- Updated
- Mar 20, 2026
- Vendor
- Bag of Words
- Product
- Bag of Words
- Classifications
- CWE-707, CWE-74, T1059.006
- Attack vector
- network
- Privileges
- authenticated
Timeline
How it unfolded
- Mar 20, 2026CVE publishedPublication date reported by the CVE source.
- Mar 20, 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=0.0.297
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
- Bag of Words generate_df command-execution proof of conceptproof 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