LangAlpha: Indirect prompt injection leading to shell and code execution
Summary
LangAlpha through 2026.09.07 contains an indirect prompt injection vulnerability. Instructions hidden in web content that the agent retrieves can make it run shell commands or Python code in the workspace sandbox without human approval. An attacker can use this to exfiltrate vault secrets stored in _internal/.vault_secrets.json, and scheduled automations can be exploited without a user present.
Affected
- ginlix-ai LangAlpha, versions 0 to and including 2026.09.07
Fixed in
Unknown.
Workaround
- Gate the Bash and ExecuteCode tools behind human approval.
- Restrict the sandbox's outbound network access and move vault secrets out of the sandbox.
- Consider disabling web fetching and scheduled automations until a fix is confirmed.
Exploitation
- No confirmed in-the-wild exploitation in the sources reviewed.
- CISA SSVC lists exploitation as 'poc' (proof of concept), with automatable 'no' and technical impact 'total'.
- No public exploit code was found during research.
Our recommendation
- Check whether LangAlpha is deployed, especially self-hosted or internet-exposed instances.
- Disable scheduled automations and public web fetching where they are not essential.
- Add human-in-the-loop approval for shell and code execution tools.
- Limit secrets in the sandbox and restrict its outbound network access.
- Monitor the upstream repository for a fix and apply it as soon as it is released.
Public exploits
PoC-in-GitHub, Metasploit and Exploit-DB list nothing for this CVE.
Research sources
- Lasso: CVE-2026-109440 www.lasso.security
- VulnCheck advisory www.vulncheck.com
- LangAlpha agent.py at referenced commit github.com
- HOL: CVE-2026-109440 hol.org
- Strix: CVE-2026-109440 www.strix.ai
- VulDB: CVE-2026-109440 vuldb.com
- OpenCVE: CVE-2026-109440 app.opencve.io
Researched and written up in English and German by an AI (Claude by Anthropic) based on NVD, CISA, EPSS and the sources listed. The vendor information is authoritative. Researched on 1 Nov 2026, 01:38.
Original description (NVD)
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.
NVD references (3)
Sources
The tracker combines official, freely usable data sources – updated continuously:
- NVD (NIST) – CVEs, rating (CVSS), affected products
- CISA KEV – vulnerabilities confirmed by CISA as actively exploited
- CISA Vulnrichment – CISA assessment (SSVC) in the official CVE record
- FIRST EPSS – probability of exploitation
- Microsoft MSRC – Patch Tuesday: vulnerabilities, severity, updates
- PoC-in-GitHub – public proof-of-concept repositories
- Metasploit (Rapid7) – modules of the Metasploit Framework
- Exploit-DB (OffSec) – archive of public exploits (links only)
- Patch days: official vendor pages (linked per vendor) (Vendor patch days)
This product uses data from the NVD API but is not endorsed or certified by the NVD. EPSS data © FIRST.org.
Affected? We can help.
We check whether your systems are affected, install updates and close the gap – continuously with the System Health Checker if you wish.