Skip to content
LLMMedium

CVE-2023-32786: In Langchain through 0.0.155, prompt injection allows an attacker to force the service to retrieve d…

Occurred
Oct 20, 2023
Published
Oct 20, 2023
Source
NVD CVE

Summary

In Langchain through 0.0.155, prompt injection allows an attacker to force the service to retrieve data from an arbitrary URL, essentially providing SSRF and potentially injecting content into downstream tasks.

Tags

  • CVE
  • CVE-2023-32786
  • cvss:HIGH

Classification

Category confidence: 90%
Severity confidence: 85%

Rationale: Langchain vulnerability enables prompt injection leading to SSRF, affecting LLM applications.

Classifier: Qwen/Qwen3.5-35B-A3B-FP8

Visit source