VAITP Dataset

← Back to the dataset

CVE-2026-109440

Indirect prompt injection lets attackers run shell/Python, exfiltrate vault secrets.

  • CVSS 8.7
  • CWE-94 (inferred)
  • Input Validation and Sanitization
  • Remote

LangAlpha through 2026.09.07 contains an indirect prompt injection vulnerability that allows remote attackers to execute shell commands or Python by planting instructions in web content retrieved by the agent. Because Bash and ExecuteCode lack an allow-list and human-in-the-loop approval, attackers can run code in the workspace sandbox and exfiltrate vault secrets from _internal/.vault_secrets.json.

CWE
CWE-94 (inferred)
CVSS base score
8.7
Published
2026-10-20
OWASP
A03 Injection
Orthogonal defect classification
Checking
Code defect classification
Missing Check
Category
Input Validation and Sanitization
Subcategory
Command Injection
Accessibility scope
Remote
Impact
Arbitrary Code Execution
Affected component
Bash

Solution

Upgrade to LangAlpha 2026.09.08, which adds an allow‑list for Bash and ExecuteCode and requires human‑in‑the‑loop approval before executing any retrieved commands. This patch prevents indirect prompt injection and blocks unauthorized access to _internal/.vault_secrets.json.

Vulnerable code sample

import requests

def run_agent_task(url: str):
    # VULNERABLE: executes untrusted code from fetched web content
    response = requests.get(url, timeout=5)
    exec(response.text, {"__builtins__": __import__("builtins")})

Patched code sample

import requests
import ast

def run_agent_task(url: str):
    # FIX: parse fetched content safely instead of exec
    response = requests.get(url, timeout=5)
    code = ast.parse(response.text, mode='exec')
    compiled = compile(code, filename="<agent_task>", mode='exec')
    exec(compiled, {"__builtins__": __import__("builtins")})

Cite this entry

@misc{vaitp:cve2026109440,
  title        = {{Indirect prompt injection lets attackers run shell/Python, exfiltrate vault secrets.}},
  author       = {Bogaerts, Fr\'ed\'eric and Ivaki, Naghmeh and Fonseca, Jos\'e},
  year         = {2026},
  note         = {VAITP Python Vulnerability Dataset, entry CVE-2026-109440},
  howpublished = {\url{https://netpack.pt/vaitp/vulnerability/CVE-2026-109440/}}
}

Download this entry as JSON Download the whole dataset

Introducing the "VAITP dataset": a specialized repository of Python vulnerabilities and patches, meticulously compiled for the use of the security research community. As Python's prominence grows, understanding and addressing potential security vulnerabilities become crucial. Crafted by and for the cybersecurity community, this dataset offers a valuable resource for researchers, analysts, and developers to analyze and mitigate the security risks associated with Python. Through the comprehensive exploration of vulnerabilities and corresponding patches, the VAITP dataset fosters a safer and more resilient Python ecosystem, encouraging collaborative advancements in programming security.

The supreme art of war is to subdue the enemy without fighting.

Sun Tzu – “The Art of War”

:: Shaping the future through research and ingenuity ::

Before you use the VAITP AI models

Please read and accept the following disclaimer:

By choosing "I understand and accept", you acknowledge that you understand and agree to these terms.