CVE-2026-109440Published Oct 20, 2026 · Oct 20Updated Oct 28
LangAlpha sandbox code execution via injected web content
ginlix-ai LangAlpha
High8.7
Affected versions
| Product | Affected | Fixed in |
|---|---|---|
| ginlix-ai LangAlpha |
| No fix yet |
Original description
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.
From the CVE record · VulnCheck