RCE Vulnerabilities in LLM-Integrated Applications via Prompt Injection

# RCE Vulnerabilities in LLM-Integrated Applications via Prompt Injection

https://gcdradar.vercel.app/incidents/rce-vulnerabilities-in-llm-integrated-applications-via-promp · AIR-2026-0350 · closed 2026-04-14

- Lane: Research
- Severity: 1 (Research / proof of concept)
- Fix status: Fix status unknown
- First reported: 2026-03-31 (happened 2025-02-27)
- AI role: Attack on AI
- Products: LangChain, LlamaIndex, Langflow, Auto-GPT

## Summary

Researchers identified 20 remote code execution vulnerabilities across 11 different LLM frameworks. The study showed that applications built with these frameworks could be exploited using prompt injection.

## What to do

No action recorded; read the sources below and the vendor's advisory.

## Latest updates

- 2026-03-31: LLMSmith Website (https://sites.google.com/view/llmsmith)
- 2026-03-31: Demystifying RCE Vulnerabilities in LLM-Integrated Apps (https://arxiv.org/abs/2309.02926)
- 2026-03-31: LLMSmith: RCE Vulnerabilities in LLM-Integrated Applications (https://atlas.mitre.org/studies/AML.CS0052)

## Sources

- 2026-03-31 MITRE ATLAS: LLMSmith: RCE Vulnerabilities in LLM-Integrated Applications https://atlas.mitre.org/studies/AML.CS0052
- 2026-03-31 arXiv: Demystifying RCE Vulnerabilities in LLM-Integrated Apps https://arxiv.org/abs/2309.02926
- 2026-03-31 sites.google.com: LLMSmith Website https://sites.google.com/view/llmsmith

Verify against the vendor's advisory before changing production. Data CC BY-SA 4.0, https://gcdradar.vercel.app.